While, in theory, the telephone company (claims it) can change your number if there is a service problem or - again in theory - for any reason it wants to, with the ability of the customer to move the number to a different Local Exchange Company (LEC-Incumbent ILEC or Competitive CLEC) or even move numbers between wireless and wireline companies, Verizon's claim that the customer does not 'own' the phone number is specious at a minimum and in any case clearly incorrect.
We as customers are paying a few cents every month for local number portability. If the customer can move their number without consent of the carrier - which is the case - then the carrier's claim that the customer does not own the number is clearly incorrect.
When competitive local (wired) phone service came into Maryland about 5 years ago I moved from (what was then) Bell Atlantic to Starpower faster than you can say 'Long distance is the next best thing to being there.' All (what is now) Verizon could do was send me a final bill and wish me a goodbye.
Since the carrier no longer has any means to refuse to let you change carriers - even if you still owe them money - then obviously the carrier no longer owns the phone number, the subscriber does.
When I moved to Virginia, I signed up with Starpower directly and have had service with them for three years. (Yes, I know the actual service comes from Verizon but it's at least a partial victory.)
The point remains, if I can take my (wireline) number from Verizon to someone else - or have a number issued from a different carrier and move it elsewhere - then obviously I own that number. This was the standard for 800/888 etc. numbers for years now, has been the standard for wired customer numbers and is now established for cellular numbers too.
It will probably take a tedious suit against some telephone company to establish de jure what is already de facto: that the customer now owns the phone number, not the carrier.
----
Paul Robinson <Postmaster@paul.washington.dc.us>
Remeber if you work as a level designer in a company making a game [don't] expect to work with tools that are full fledged.
The tools you are going to work with will constantly evolve, your tools availble will be added and removed as the game moves on.
I shudder to think, with the level of the tools that are being distributed with the published editions, how bad the stuff that is used internally (and isn't even up to release quality) is like. Having used the third-party one done for Doom, and the one included with Duke Nukem, I'd say there isn't a dime's bit of difference whether it's internally or externally developed, they're all pretty much the same: they don't do all of what you need and they're full of bugs.:)
I've had 27 years of experience programming and I can't figure out how to use Q3Radiant (the editor for Quake III Arena). Maybe I'm just not that bright.:)
Quake 3 comes with the same editor that the id team used (and designed) and it's free.
Quake 3... Use it... Love it...
I would love to try doing a level with the editor for Quake III. I've been a programmer for 27 years, from PCs to micros to mainframes, and going back when you used punched cards to enter programs. I cannot figure out how to use the editor. And I've done maps in Doom, in Duke Nukem and in other games, but the editing program is just too complicated for me to figure out how to get it to work. Maybe I'm just not very bright.:)
I have found, whether the editing tools are written by third parties unrelated to the developers (as the ones for DOOM were) or were released with the game (Duke Nukem 3d, Quake 3 Arena) the tools generally suck.
They'll get the job done (presuming you can figure out how to use it; more on this below), but often there are severe bugs that damage the map, you have to remember to do alternate saves under different names (to recover from this problem), you have to test the map in another program from the one you're designing in, etc.
The method for designing maps usually involve designing an area, then perhaps setting parameters such as size, height above or below 0 of the floor, lighting levels, etc.
The problem begins because they use a graphic-based system to design maps, there is no capability to edit a map in some form in which you could specify parameters as text in order to check. Often the display gives you no information about the area in question except a visual image. You have no way of figuring out what the dimensions of the area are.
Sometimes the game editor will crash for unexplained reasons. Sometimes it will inexplicably damage a map without having any way to reset it back, essentially destroying the work you've done.
Then there is the complexity of the tools. Up through Duke Nukem it was reasonable to be able to edit a page, but the new tools for Quake III for example, are impossible to figure out how to use.
Maybe I'm just not very bright, but I don't think I'm all that stupid, I'm 43 years old and I've been doing programming for 27 years, on everything up through and including mainframes.
When you want to enable features in these editors, it usually requires including some kind of flag sprite or other image, perhaps on an image if you want to make it an actor, then setting parameters on that sprite in a different mode.
Often you need to use special textures or color modes to enable certain features, but it's not made clear when this is necessary.
Whether this is intentional to sell people a reference manual on the game editor, or there just isn't much interest in documenting a secondary tool, there is usually no help and no explanations available about how to use these tools, and often no explanation on what to do with them.
what if i turn somebody that just recently cleared their HD of everything, and say that they hid themselves really well and didn't use their own computer to spread the worms...
Usually these type rewards are based on either 'arrest and indictment' or 'arrest and conviction' so unless there is at least enough information present for them to present a grand jury indictment or for the guy to actually be convicted, you get bupkes.
They only offer the money if the person who wrote it is arrested. If these worms originated from China, Brazil or Indonesia, I'm pretty certain no-one is getting arrested.
For that kind of money, don't be surprised if someone doesn't 'accidentally' take a quick unscheduled exit out of that country, with or without a passport. H. Ross Perot successfully got his own people out of Iran years before Carter botched the attempted rescue of the American Hostages; and when a DEA agent was murdered in Mexico, the U.S. government kidnaped Doctor Alvarez-Machain, a Mexican national, to be tried in the U.S. for the alleged murder which was committed in Mexico. The original trial court said this was an illegal kidnapping in violation of the extradition treaty between the two countries. The Supreme Court said that despite there being an extradition treaty, grabbing him that way was okay. What was embarassing was he was later found not guilty anyway.
Being in some foreign country that doesn't have a law against it won't protect someone for long if someone who was affected by it is mad enough to bring them to a country where it is. (The courts consistently have said they will not consider how the person got to them, whether the person voluntarily walked in or a private party kidnaped them and brought them to court.)
Some people have done this already. It's called a Honeypot. You put up a machine either with nothing of importance or with faked material in order to attract people to try and break into it for the purpose of figuring out how they are doing so.
Paul Robinson <Postmaster@paul.washington.dc.us>
That is a damn shame, too. We have enough trouble keeping programming jobs here, and Microsoft has to go and think of another way to make it harder to develop software that they don't make money off of.
costing Microsoft an estimated $40 million to $50 million a year.
It's not costing Microsoft jack, because that $40-50 million never existed.
The above poster has it right on the money. For some reason, when counting losses allegedly due to piracy, it's presumed that every single copy that was not licenced would have been a 'list price, full retail purchase' and thus overinflates the amount of piracy (which makes draconian solutions like DMCA and 'Trusted Computing' look necessary to stem those 'huge losses') when the actual number of actual losses of real money in the form of sales that were not made, which would have been made if the unlicensed copies had not been made, would certainly have been much less. Paul Robinson <Postmaster@paul.washington.dc.us>
Was that at the end of the article (about textbooks selling at a discount from the manufacturer's sticker price) were sponsored links for places that sell textbooks at a discount and places that buy and sell used textbooks!
Paul Robinson <Postmaster@paul.washington.dc.us>
The work being misused is on a CD and not in terms of the license. DMCA applies generally to works on-line and not to a work not on a computer. Arguably the use of the table of contents could be considered fair use even if used differently from what your license says.
You have a book. It is copied (not as a copy on a computer system) other than in a method you approve. It is not a software program that was copied, nor is it an access control system that was defeated. What you have is a "moral rights" violation under the Berne Union Copyright Treaty and U.S. Law. I do not think the DMCA applies here.
Paul Robinson <postmaster@paul.washington.dc.us>, or <paul@elusive-butterfly.net>
Notice that U.S. law requires companies to refuse to do business with, i.e. to boycott Cuba, Libya, North Korea and a few other countries.
U.S. Law also makes it illegal to cooperate in any boycott against Israel by other countries.
I wonder how the U.S. would take it if other countries prohibited U.S. companies from doing business there if they cooperate in boycotts against Cuba, North Korea or those countries, or required them to comply with their anti-boycott rules if they prohibited such practices?
Not that I personally agree with how those countries operate, I find it rather amusing that the U.S. requires a boycott of certain other countries, but makes it illegal to follow required boycotts by some countries of their disfavored country(ies).
Sorry, but if you're going to contribute to a community, you have certain obligations to fill
No. You don't. It's FREE software. "Free" means that it doesn't take rights away from the user, and it doesn't impose additional rights on the developer. What else would you think "free" meant?
If everybody makes an obscure interface for their app just so they can make a few bucks on a manual, then who's going to adopt it?
THEN USE SOMETHING ELSE!!!!!!!
I could not agree more. You don't like it, take your (lack of) business for them and (nonexistent) money elsewhere. If people do not like that they have an NDA for the documentation, they won't use it and they will learn. If they can make a living this way, it means they were right. Wasn't the original UNIX source code released this way?
Paul Robinson <postmaster@paul.washington.dc.us> http://paul.washington.dc.us
Write some code for your new app. Write a code obfuscator; a program that takes some C code and produces some more C code that's functionally equivalent but virtually unreadable to humans.
Sometimes the code people write as is without an obfuscator is all but unreadable to start with, but that's beside the point!:)
(Even changing all identifier names to foo1, foo2 would make it difficult enough to figure out what the code does. Of course, far more obfuscatory obfuscations are possible.)Finally, run the code for your app through the code obfuscator and publish the resulting code under the GPL, claiming it to be the original source.
They don't even have to do this. They can ssy this is what they are going to release. It's their choice. You don't like it, don't use it or try and negotiate other terms. Everything is or should be negotiable.
Charge to see the documentation and include a NDA.
That's their choice. Even if the source is obfuscated, it's still possible to learn some thing from it. And if that becomes common, I suspect the next thing would be de-obfuscators which would clean up the code to make it more readable. In fact, that's the function of reverse-engineering tools developed to fix Y2K bugs in old source code lacking documentation.
Isn't there at least a moral imperative to publish readable source code under the GPL?
Why?
Who said they owe you or anyone else anything related to the software they have developed? By what right? They don't even have to release anything at all. You don't like it, don't use it.
Nobody owes anyone else anything with one exception:
"It is what you owe me..."
"I don't owe you anything. You chose to bring me into this world; I had no choice to come here. You owed me the means to live until I became an adult. This was something you owed me as a matter of right. As it will be, when I have them, what I will owe to my children, that I will owe to them as a matter of right. I owe you nothing."
- Sydney Poitier to his father in "Guess Who's Coming to Dinner"
Unless you're their child, they don't owe you anything.
The problem was the implication that once you'd learned how to solve your problem you couldn't divulge the information to third parties, in other words help others.
That is their right. You don't like it, use the source code without using their documentation.
Not only is this against the spirit of free software, it can arguably be a violation of the GPL:
The GPL is a license for the party that wants to use the source code of the software. The GPL is not applicable to the party licensing the sortware, they can do anything they want, including offering you a license on different terms. And the GPL doesn't apply to anything that doesn't explicitly include it. Besides that, if I'm not mistaken, you're not subject to the terms of an NDA, even if you signed it, if you can show you legitimately got the information disclosed as a result of material you got under the NDA from some other place without the restrictions of that NDA.
Suppose you fork the project, but couldn't have written your mods without reading the documentation. Is this a violation of the NDA?
That's the sort of question that makes lawyers rich. While I do study law because I need to know what I can get away with, I'm not a lawyer, and because I haven't studied the law regarding NDAs I can't say, but I suspect quite possibly.
Or more bluntly, if for all practical matters the license terms forbid you to post a makefile if you've read the documentation before writing it, is this really free software?
The question is, can you access the sourcecode without restrictions? (By restrictions I mean do you have to do anything beyond that required by the GPL) If you can, well, there's the source code, use it without the NDA or get the documentation and agree to the terms of the NDA. If you're not paying them anything, they owe you nothing.
"It is more accurate to say that a license is a promise not to sue someone for infringing your rights." - Robert A. Kreiss, University of Dayton School of Law
This may be one way the vendor has figured a way to release the product as open source but still make money off of it. There is nothing that says they have to create any documentation about it at all.
In fact, some people have suggested to the company that they work for, that in view of how i makes the vast percentage of its revenue charging for configuring, setting up and maintaining the software they develop, that it might be worth it to open source the product and give the software away, but charge for everything else. So if you're really broke you can get it for free, or if you want to look at it before putting it on your machines, you can do that. But if you want anything else, even instructions on how to install or use it, you have to pay something.
Why is this such a problem for you? Is it that you think they should give everything away? They have to eat, too. If you don't like it, don't agree to their terms. Since the source code is available without restrictions, take the time and effort to study the code and learn how it works, pay someone else to do that, or pay them and/or agree to their restrictions.
Has anyone noticed there are no open-source tax preparation or payroll software programs? (If I am wrong, someone e-mail me). Because those aren't very sexy for programmers to work with - which means that people aren't volunteering to do them for free - and because they require constant maintenance. (Not [merely] because of bugs, but because the tax laws and payroll processing rules change every year.)
Someone has to pay for the maintenance since this is not something your average programmer either wants to do for free (in the case of a payroll product) or has the resources to do on his own (in the case of a tax preparation program.)
Now, I know that there is GNU Cash as a workalike for Quicken but I know of no open-source software for mundane apps like payroll or tax preparation, and if there were, I can't see how we could expect them to be kept up to date without significant resources to handle the average of 10,000 tax law changes yearly. And that's just the U.S.
Every country has its own rules and thus a tax package to handle the U.S. Internal Revenue Service (IRS) rules would be worthless for Canada Customs and Revenue Agency (CCRA) or for the UK's Inland Revenue (IR). Or the other tax authorities in the other 160+ jurisdictions around the world, almost all collecting some form of income tax. Then there's the 30+ states in the U.S. that also impose taxes on income, provinces in Canada (if they do, I'm not sure) and other subdivisions of governments elsewhere.
Now, some of these agencies are providing on-line tax preparation over their websites, but the method to do this is not open source, and would you expect to pay the lowest possible tax by using, or would you really want to trust, a tax package developed by the taxing authority?:)
Historical note: the typical quote "The power to tax is the power to destroy" was originally written by U.S. Chief Justice John Marshall as "That the power of taxing by the States may be exercised so as to destroy..." McCulloch v. Maryland4 Wheat. 316 (1819), the first case declaring a statute void for violating the constitution.
I have thought of this on occasion. I think I was at most 6 when I saw the first thing that I can remember.
I think the first thing I remember was the stars on a bottle of soda - cream soda I do believe - these were 4-point stars, I distinctly remember that. I remember them as white paint on a clear glass bottle.
As a programmer I would say that was when I was instantiated. (As a "child" process. Pun intentional!)
As a philosopher I would say that's when I remember coming into existence, because while people tell me about things I did and said before that, I have no recollection of them.
A friend of mine, whom I have never met, who lives about 1,000 miles away, says that she can remember things as an infant. That amazes me.
I thimk I had to be 5 or 6 as far as when my earliest memories occurred (middle 1960s) because some time later, which was long enough for me to have memories of being in the city where that first memory happened for some time, so I would say it was probably at least a year or more later, our family moved from Wisconsin to Miami and I remember living in a trailer park there. At the time I was there I was about 7 so that's how I date the first memory I ever had.
I remember how there was a little kid there in the trailer park, a neighbor at another trailer - little kid meaning he was younger than I was, probably no more than 4 or 5 - and how I would make him laugh by using swear words. The thing was, he knew they were swear words, which is why he would laugh at them, and while I kind of knew that they were classed as such, it would not be until some time later - maybe years, I'm not sure - that I would be able to "feel" that they were and recognize them as such.
So knowing that event happened when I was 7 allows me to date the first memory of my existence at least a year or two before that.
Paul Robinson <postmaster@paul.washington.dc.us> http://paul.washington.dc.us
Among other things, "A Philosopher, Computer Programmer, and Notary Public in and for the Commonwealth of Virginia."
"Above all else... We shall go on..." "And continue!"
The flaw in your logic is pretty easy.
You don't run a business that is BASED off a database on retail quality IDE drives you buy at bestbuy.
I wouldn't buy anything I expected to use in a commercial environment at a store that is not devoted to computers because they won't have the selection and the prices will probably not even be competitive.
Were talking high speed SCSI drives.. in a type 5 raid at LEAST ! if not redundant raids.
I believe I did give an example of dual mirrored drives. So go quad mirror.
On a professional development platform (the kind you have to pay for.. not MySql) Backups, system admin salaries, DBA salaries etc.
You don't run a game that is expected to get over a million subscribers in the first 3 months, on your brothers old 486 tripped out to run linux with a spare parts from your friends.
Well, that very answer shows that his whining is even more pedantic. I did not realize the operation was that large. But as the numbers rise, so do the scale factors. Bur either you put in the correct amount of capital to build the business in the first place, or you start small and work up. You don't underfund the system, refuse to use some of the income from the business to feed it - which is what a professional business manager does - and then expect a business that is a thriving operation be able to continue operate when you starve it of the cash that it needs to grow that it itself produces. This is on the order of a farmer borrowing money, starting a crop, picking it, paying back the loan but refusing to use part of the money the crop produces to pay for the seeds you need to buy to grow the next crop, then whining how he couldn't afford to grow another crop
If this is supposed to produce a million customers over a three-month period, either it produces something in the neighborhood of $10 million a month if it charges out at $10 a month, or it produces in the neighborhood of $10 million a year if it's $10 a year. All up-front cash since the customer pays when he signs up. For that kind of money, either build the place to support the load or close down, refund the customer's money and stop complaining. You don't underfund the necessary equipment to run an operation then complain how you can't run it successfully on less than was necessary to operate it in the first place. That's whining and that's exactly what he's doing. Do it right, make it right, or admit you can't and close up.
Not when every single MINUTE of downtime costs you money (just in Customer Service costs alone.. let alone lost revenue and PR to counteract the negative experience.)
Well, fine, we're talking about a business that should conceivably return several million a year in profit after deducting costs. That means that they should expect to put in the investment necessary to support the operation, or don't expect to get that big that fast and don't spend as much. Do one or the other. Anything else is dishonest.
48,000 customers won't even cover the number of copies shipping to california i bet:P
At the time I did not know the sizes involved. So raise the cost estimates I give in my other example by a factor of 20, and you get $60,000 for disk space for three years. And realize you don't need it all at once, when you drop below 50% free you buy another bunch of boxes or whatever. But I believe my original point is correct. Besides that, if it's that large an operation, either it should be spun off as a separate company where that is the only thing they do, or it should probably be operated on a managed hosting system which can probably handle the care and feeding of servers better than they can. Do something correctly or find someone who can, don't do a half-way job and then complain about how expensive it is.
Of course.. Ralf's point about service size (which is only a SMALL part of this decision) is scaleability. They have to project over a 5-7 year period. and growth in that time.. So.. conservitaly , if they are expecting 1 million accounts in the first year.. they could concievably need space for 4 million accounts over a 7 year life.
If you have ever worked in a professional big business position.. you know how hard it is to get funding 1/2 way through a project - no matter HOW important or sucessful it has been.
I guess it is this sort of dishonest way of operating projects is the reason more than 1/2 of all software projects get cancelled. Or is it 3/4? Improperly funding a project and incompetent management are the usual causes of that sort of failure. I've never worked in a big business because whenever I've applied at one of those places, the usual comment was I was overqualified for the position. There were a number of times I wanted to paste someone in the mouth for insulting me that way. To me, saying you won't hire someone because they are too good is saying that they want nothing but incompetents and are afraid of people who know what they are doing. Maybe that's the reason the manager of this system is whining instead of doing it right in the first place.
People who always say "well IDE drives are cheap, just buy a cheap computer and stack it with drives" proves they have never put together or been around real servers.
People who think you should just throw money at a problem have never tried to run a sustainable business. The correct answer is, you start small, as the business grows and produces an income stream, you use the income from the business to grow the business which includes improving the equipment as you make money off of the business. Anyone can run a business if you have lots of cash to burn... until you run out of cash. If you start out with very little money and make the business pay for itself, you can stick around for the long haul. Because your risk is lower, your upside is much higher and your downside is substantially reduced. That's why Real Estate investing beats anything else. When someone uses leverage of someone else's money, the return rate can be as high as infinite.
That may be fine for your little box sitting in your apartment that you use for ftp and telnet when you are away or serving the occasional web page, but for real life servers that actual need high performance and long uptime that just won't cut it.
My example focused exclusively on the use of additional boxes for disk space only which was the issue the man on the game system was complaining about. I was not saying anything about what was being used for the actual game server, just as a system to provide additional external disk space separate from the server machine over a separate intranet connected to it.
In my example I proposed the idea of running dual mirrored machines as RAID boxes. Are you trying to tell me that two identical machines running on an operating system that, according to the published reports, provides uptime reliability measured in years between reboots other than for planned outages is inadequately reliable for most transactional based systems? Then make it 4 machines, or 6 or 8 or whatever is reasonable.
There are two ways to increase reliability. One is to harden the system to prevent failure by building expensive boxes of extremely high reliability, with redundant components, dual processors, dual power supplies, etc. The other is to reduce the complexity of the system so that any particular failure at any point does not compromise the system.
Maybe I am wrong, but let's consider the cost of setting up a system consisting of 16 cheap computers each on a separate UPS, each containing a single inexpensive 60 billion byte drive and a gigabit network card, so that the system represents 4 of each drive group mirroring all the data sent to that group, doesn't that then provide the equivalent of 4 x 60 billion bytes, or 480 GB of storage, with at least four times the reliability (since you now have to have 4 boxes all simultaneously fail before any one "drive" is out of service, and each one only has to respond to 25% of the read requests for that group) at a fraction of the cost of a more expensive system that it in and of itself supposedly more reliable? Further, I think it's even more reliable since you can conceivably continue to operate, even in degraded mode, until 15 of the boxes fail which I do believe the possibility of failure only grows linearly while reliability grows exponentially as more machines are added.
A specific example of this is in the use of birth control. Any of the usual methods, pills, barriers, or chemicals, might have a 1-3% failure rate, i.e. 3 in 100. So, if a woman uses two forms of contraception, say diaphragm and foam together, the worst failure rate goes from 3 in 100 (3%) to 9 in 10000 (0.009%). If she can talk the guy into also using a condom, the failure rate is now 27 in 1,000,000 (0.000027%). Increase the number of things that have to fail to cause catastrophe and the reliability rate goes up.
I was giving a gross example where you can do the job with inexpensive equipment. So I could be wrong on some of the figures. So maybe it's a bit more expensive. But I do believe you can do this sort of thing successfully with the kind of setup I was talking about for less than $5,000 in any case. And remember, what I talking about here was moving the disk space to inexpensive mirrored servers, I said nothing about the transaction server(s).
So if you wouldn't recommend Linux (or BSD or any of the other similar items) what would you recommend? I would like to learn something here if I am wrong, please explain to me how this isn't going to work. I would like to know because I do want to understand and as I understand, this seems to be the most cost effective way to do this. I could be wrong and I'd like to understand more.
"If it takes as much as 10 megabytes per user (and that's a hell of a lot of space to store character data, and probably isn't anywhere near that much), you can host 48,000 customers with mirrored disk drives for a one-time cost of about $2K."
Below I will comment on how I was using the phrase "character data" in my article and the way I phrased it, I may have caused you to misunderstand me.
Again, obviously not understanding how a real life server works. If you want quality and something that will scale really well you need to pay for it, Linux ain't it.
I was under the impression that it does scale. If you feel that Linux is inadequate to handle the scaling factors for a large database system such as a game would use, tell me what you think such an application requires. Considering that I've heard that large clustering systems of thousands of nodes use Linux successfully to handle the load, and IBM has it on S/390 mainframes which are designed to handle huge transaction volumes and conceivably run thousands of Linux servers on one machine, I believed it did scale quite beautifully. But tell me so I can learn something if you think I am wrong.
Database tables can get rather large when you add in indexes and linking all the tables together. Its not just a text file list.
My apologies. When I said "character data" I meant the data to store the values of a character in a game, not character data as in text. I realize that the way I said it could have mislead you and I apologize for the way I said it as I realize now how it could have been misunderstood.
I've worked with databases where the base system, with no data, just empty tables and stored procedures, ran in excess of 100 megabytes. I am saying that I think a factor of 10 megabytes average space per user is a lot of space. But maybe I'm wrong so let me see. If you figure a index key per record at being perhaps 16 bytes, and maybe you have 500 tables, and each table perhaps uses an average of 3, no let's say 5 links per record, and lets say the user has, oh, say, 50 records each in the system, that's 500x50x5x16 means you need about 2 megabytes for the indexes for each user. If you use normalized databases you might use maybe 2K per record based on trying not to repeat data except where needed for performance, so you need 5000K or 5 meg for the user's data. Add into that perhaps 3 meg for image data, and you have the 10 megabytes I estimated as a lot of space.
Okay, then, I was wrong on it being a lot of space. Make it 20 meg average per user and cut my estimates in half or double the cost.
If I am wrong on some of these numbers, please enlighten me so that when I make a statement in the future I will have the correct numbers and be able to justify my estimates as I am trying to do now, using my best estimates from my own experience in programming and database management. I know I do not know everything. But I can make educated guesses and perhaps, if someone else knows more than I do they can give me information to allow me to learn where I have made assumptions not backed up by real-life data.
And I wasn't figuring that to be a lifetime cost, I was figuring that based on the estimated 3 year life-span that an IDE disk is expected to have. If you have 48,000 customers each paying $10 a month to use the service - which is what I think many of these are doing - that's $480,000 a month, which is a lot of money. Based on usual cost figures of 50% for labor cost, this gives you $240,000 a month for salaries, $24,000 a month for G&A, (general and administrative ovehead) and perhaps another $48,000 a month for maintenance and equipment, and maybe another $12,000 a month for the Internet connections over multiple . This leaves in the neighborhood of $324,000 in costs which may be high. That means there is, or should be, something in the neighborhood of $100,000 a month to pay back the investors. As there are more people taking the service, it produces more income.
If the service is only being paid $10 a year, then you can run it through automated sign ups and buy managed hosting, and increase the amount of hosting as more people use the service.
The point I am making is that if you want to run a successful business you run it as an investment. You put some money in as initial capital, in the amounts necessary to get it started, according to what is reasonable to make it operational, and you expect it to eventually return cash in order to build up to larger capacities, and if it can't, you either make it pay or you close it. You do not complain that the costs of running the business make it non-economic. Then you're just whining, and as far as I can see that's exactly what he is doing, whining about how much it costs to run the operation instead of working with what you can afford and building it up as it produces an income stream.
To me it sounds suspiciously like it was expected they could just throw some money into this - probably less than it should have had - and expect it to have immediate huge cashflow and huge returns, and when it didn't they were disappointed. It don't work that way. A successful investment means you start it small, and allow the business to grow itself.
I think someone once said that those who become emotional over their investments soon end up losing money over them. And his sort of whine sounds suspiciously like emotionalism.
Paul Robinson <postmaster@paul.washington.dc.us>
I read part of the guy's letter to the people using the game. The general tone sounds like someone whining about how much the overhead costs where the overhead is part of the cost of doing business. Like complaining how much disk space costs to run these characters. If you can't afford to run the game, quit, or charge what it costs, but don't complain that it's too expensive to operate.
I am reminded of the TV show, The Flintstones, when Fred has gotten a part in a movie playing the creature from the tar pits, and they're supposed to have safety equipment so he isn't hurt. The director has a scene where the hero uses a club on the creature, and wants a padded club for the scene. The property manager whines, "Are you kidding? Do you know how much padded clubs cost?" So they'll use a real one. You can guess the result: Fred is knocked cold on the first swing to the head. It's cheaper to injure extras than protect them.
If it's that expensive they perhaps should run their own server, either renting it or collocate a box of their own and put in enough disk space. If a T1 for network traffic costs about $500 a month, disk space is peanuts in comparison.
Last time I checked, a 120 GB - that's 120,000 megabytes - IDE disk sells for about $300. Buy an inexpensive used computer, say a 200 MHZ pentium for about $50 with a $75 Gigabit network card connected to the server via a $20 crossover cable with the same $75 gigabit network card so you can access disk space fast. So if you have a box with 4 of the 120 GB space, you can run 480 GB of space for all the characters - that's 480,000 megabytes - for $1500. Or you can mirror it and have 280GB of space completely mirrored. Or spend perhaps another $500, and put two drives each in separate machines with a gigabit switch between them. Or maybe the drives aren't that fast and you can get by with $15.00 100 megabit networking cards and a $20 hub. (You put the traffic going to and from the external disk drives on a separate network card so the data traffic doesn't interfere with the network traffic which is playing the game.)
If it takes as much as 10 megabytes per user (and that's a hell of a lot of space to store character data, and probably isn't anywhere near that much), you can host 48,000 customers with mirrored disk drives for a one-time cost of about $2K. And I believe all the software to handle this including raid striping or mirroring is built in to Linux which means it's free. If your provider is charging too much for disk space, colocate and reduce your cost. Or charge what it costs to operate the game.
This is just another way that companies that operate on-line games get greedy because they don't want the customers figuring ways to make money off the game (by selling characters they've built up) or want to impose draconian conditions (like no fan sites or no creating mods to their game) because they think the customer and any money he spends related to this product (and any attention he devotes to it) is all theirs and nobody else should be able to make any money at all related to it unless they get it or a huge percentage, or any non-company attention unless they control how that attention is given.
Anyone notice how when movies are released to theatres, the studio gets 90% of the ticket price? (The only thing keeping movie theatres alive is the popcorn stand.) Also notice how many studios, when they release films or tv shows where a star in it took a smaller salary in exchange for profit participation, discovers after millions of dollars in sales, that the show or movie has always lost money and thus allegedly never had any profit to pay out? I think it's no coincidence that some of these on-line games are run by divisions of studios and they are operating them the same way as the studios are run. They think everything related to their properties should be all theirs and resent anyone getting anything out of it unless they get a piece of the action. No, make that, they get as much as possible of the action, and resent sharing anything they might have to allow.
Slashdot Mirror
While, in theory, the telephone company (claims it) can change your number if there is a service problem or - again in theory - for any reason it wants to, with the ability of the customer to move the number to a different Local Exchange Company (LEC-Incumbent ILEC or Competitive CLEC) or even move numbers between wireless and wireline companies, Verizon's claim that the customer does not 'own' the phone number is specious at a minimum and in any case clearly incorrect.
We as customers are paying a few cents every month for local number portability. If the customer can move their number without consent of the carrier - which is the case - then the carrier's claim that the customer does not own the number is clearly incorrect.
When competitive local (wired) phone service came into Maryland about 5 years ago I moved from (what was then) Bell Atlantic to Starpower faster than you can say 'Long distance is the next best thing to being there.' All (what is now) Verizon could do was send me a final bill and wish me a goodbye.
Since the carrier no longer has any means to refuse to let you change carriers - even if you still owe them money - then obviously the carrier no longer owns the phone number, the subscriber does.
When I moved to Virginia, I signed up with Starpower directly and have had service with them for three years. (Yes, I know the actual service comes from Verizon but it's at least a partial victory.)
The point remains, if I can take my (wireline) number from Verizon to someone else - or have a number issued from a different carrier and move it elsewhere - then obviously I own that number. This was the standard for 800/888 etc. numbers for years now, has been the standard for wired customer numbers and is now established for cellular numbers too.
It will probably take a tedious suit against some telephone company to establish de jure what is already de facto: that the customer now owns the phone number, not the carrier.
----Paul Robinson <Postmaster@paul.washington.dc.us>
I shudder to think, with the level of the tools that are being distributed with the published editions, how bad the stuff that is used internally (and isn't even up to release quality) is like. Having used the third-party one done for Doom, and the one included with Duke Nukem, I'd say there isn't a dime's bit of difference whether it's internally or externally developed, they're all pretty much the same: they don't do all of what you need and they're full of bugs. :)
I've had 27 years of experience programming and I can't figure out how to use Q3Radiant (the editor for Quake III Arena). Maybe I'm just not that bright. :)
Paul Robinson <Postmaster@paul.washington.dc.us>
Paul Robinson <Postmaster@paul.washington.dc.us>
They'll get the job done (presuming you can figure out how to use it; more on this below), but often there are severe bugs that damage the map, you have to remember to do alternate saves under different names (to recover from this problem), you have to test the map in another program from the one you're designing in, etc.
The method for designing maps usually involve designing an area, then perhaps setting parameters such as size, height above or below 0 of the floor, lighting levels, etc.
The problem begins because they use a graphic-based system to design maps, there is no capability to edit a map in some form in which you could specify parameters as text in order to check. Often the display gives you no information about the area in question except a visual image. You have no way of figuring out what the dimensions of the area are.
Sometimes the game editor will crash for unexplained reasons. Sometimes it will inexplicably damage a map without having any way to reset it back, essentially destroying the work you've done.Then there is the complexity of the tools. Up through Duke Nukem it was reasonable to be able to edit a page, but the new tools for Quake III for example, are impossible to figure out how to use.
Maybe I'm just not very bright, but I don't think I'm all that stupid, I'm 43 years old and I've been doing programming for 27 years, on everything up through and including mainframes.
When you want to enable features in these editors, it usually requires including some kind of flag sprite or other image, perhaps on an image if you want to make it an actor, then setting parameters on that sprite in a different mode.
Often you need to use special textures or color modes to enable certain features, but it's not made clear when this is necessary.
Whether this is intentional to sell people a reference manual on the game editor, or there just isn't much interest in documenting a secondary tool, there is usually no help and no explanations available about how to use these tools, and often no explanation on what to do with them.
Paul Robinson <Postmaster@paul.washington.dc.us>Since it's a database program and now open source, I guess we can add the obligatory quote:
"For the memory of a lifetime, Rekall, Rekall, Rekall..."
Paul Robinson <Postmaster@paul.washington.dc.us>
Explain how you are going to profit when the minimum bail will probably be at least two million dollars?
Being in some foreign country that doesn't have a law against it won't protect someone for long if someone who was affected by it is mad enough to bring them to a country where it is. (The courts consistently have said they will not consider how the person got to them, whether the person voluntarily walked in or a private party kidnaped them and brought them to court.)
Paul Robinson >Postmaster@paul.washington.dc.us>- Virus writers
- Spammers
- Pop-up ad issuers
<KIDDING>Now all we need is an effective death penalty act for the worst of the above.</KIDDING>I personally do not 'hate' Microsoft, and I agree that this is definitely a Good Thing as well.
Some people have done this already. It's called a Honeypot. You put up a machine either with nothing of importance or with faked material in order to attract people to try and break into it for the purpose of figuring out how they are doing so. Paul Robinson <Postmaster@paul.washington.dc.us>
(For those that don't get the simile, I'm referring to the scene of Neo's interrogation in The Matrix.)
Paul Robinson <Postmaster@paul.washington.dc.us>That is a damn shame, too. We have enough trouble keeping programming jobs here, and Microsoft has to go and think of another way to make it harder to develop software that they don't make money off of.
Paul Robinson <Postmaster@paul.washington.dc.us>
Was that at the end of the article (about textbooks selling at a discount from the manufacturer's sticker price) were sponsored links for places that sell textbooks at a discount and places that buy and sell used textbooks!
Paul Robinson <Postmaster@paul.washington.dc.us>
You have a book. It is copied (not as a copy on a computer system) other than in a method you approve. It is not a software program that was copied, nor is it an access control system that was defeated. What you have is a "moral rights" violation under the Berne Union Copyright Treaty and U.S. Law. I do not think the DMCA applies here.
Paul Robinson <postmaster@paul.washington.dc.us>, or <paul@elusive-butterfly.net>
U.S. Law also makes it illegal to cooperate in any boycott against Israel by other countries.
I wonder how the U.S. would take it if other countries prohibited U.S. companies from doing business there if they cooperate in boycotts against Cuba, North Korea or those countries, or required them to comply with their anti-boycott rules if they prohibited such practices?
Not that I personally agree with how those countries operate, I find it rather amusing that the U.S. requires a boycott of certain other countries, but makes it illegal to follow required boycotts by some countries of their disfavored country(ies).
Paul Robinson < postmaster@paul.washington.dc.us>, or < paul@elusive-butterfly.net>
Paul Robinson <postmaster@paul.washington.dc.us>
http://paul.washington.dc.us
Who said they owe you or anyone else anything related to the software they have developed? By what right? They don't even have to release anything at all. You don't like it, don't use it.
Nobody owes anyone else anything with one exception:
"It is what you owe me..."
"I don't owe you anything. You chose to bring me into this world; I had no choice to come here. You owed me the means to live until I became an adult. This was something you owed me as a matter of right. As it will be, when I have them, what I will owe to my children, that I will owe to them as a matter of right. I owe you nothing."
- Sydney Poitier to his father in "Guess Who's Coming to Dinner"
Unless you're their child, they don't owe you anything.
Paul Robinson <Postmaster@paul.washington.dc.us>
http://paul.washington.dc.us
"It is more accurate to say that a license is a promise not to sue someone for infringing your rights." - Robert A. Kreiss, University of Dayton School of Law
Paul Robinson < postmaster@paul.washington.dc.us>
In fact, some people have suggested to the company that they work for, that in view of how i makes the vast percentage of its revenue charging for configuring, setting up and maintaining the software they develop, that it might be worth it to open source the product and give the software away, but charge for everything else. So if you're really broke you can get it for free, or if you want to look at it before putting it on your machines, you can do that. But if you want anything else, even instructions on how to install or use it, you have to pay something.
Why is this such a problem for you? Is it that you think they should give everything away? They have to eat, too. If you don't like it, don't agree to their terms. Since the source code is available without restrictions, take the time and effort to study the code and learn how it works, pay someone else to do that, or pay them and/or agree to their restrictions.
Has anyone noticed there are no open-source tax preparation or payroll software programs? (If I am wrong, someone e-mail me). Because those aren't very sexy for programmers to work with - which means that people aren't volunteering to do them for free - and because they require constant maintenance. (Not [merely] because of bugs, but because the tax laws and payroll processing rules change every year.)
Someone has to pay for the maintenance since this is not something your average programmer either wants to do for free (in the case of a payroll product) or has the resources to do on his own (in the case of a tax preparation program.)
Now, I know that there is GNU Cash as a workalike for Quicken but I know of no open-source software for mundane apps like payroll or tax preparation, and if there were, I can't see how we could expect them to be kept up to date without significant resources to handle the average of 10,000 tax law changes yearly. And that's just the U.S.
Every country has its own rules and thus a tax package to handle the U.S. Internal Revenue Service (IRS) rules would be worthless for Canada Customs and Revenue Agency (CCRA) or for the UK's Inland Revenue (IR). Or the other tax authorities in the other 160+ jurisdictions around the world, almost all collecting some form of income tax. Then there's the 30+ states in the U.S. that also impose taxes on income, provinces in Canada (if they do, I'm not sure) and other subdivisions of governments elsewhere.
Now, some of these agencies are providing on-line tax preparation over their websites, but the method to do this is not open source, and would you expect to pay the lowest possible tax by using, or would you really want to trust, a tax package developed by the taxing authority? :)
Historical note: the typical quote "The power to tax is the power to destroy" was originally written by U.S. Chief Justice John Marshall as "That the power of taxing by the States may be exercised so as to destroy..." McCulloch v. Maryland 4 Wheat. 316 (1819), the first case declaring a statute void for violating the constitution.
Paul Robinson <Postmaster@paul.washington.dc.us>
http://paul.washington.dc.us
I think the first thing I remember was the stars on a bottle of soda - cream soda I do believe - these were 4-point stars, I distinctly remember that. I remember them as white paint on a clear glass bottle.
As a programmer I would say that was when I was instantiated. (As a "child" process. Pun intentional!)
As a philosopher I would say that's when I remember coming into existence, because while people tell me about things I did and said before that, I have no recollection of them.
A friend of mine, whom I have never met, who lives about 1,000 miles away, says that she can remember things as an infant. That amazes me.
I thimk I had to be 5 or 6 as far as when my earliest memories occurred (middle 1960s) because some time later, which was long enough for me to have memories of being in the city where that first memory happened for some time, so I would say it was probably at least a year or more later, our family moved from Wisconsin to Miami and I remember living in a trailer park there. At the time I was there I was about 7 so that's how I date the first memory I ever had.
I remember how there was a little kid there in the trailer park, a neighbor at another trailer - little kid meaning he was younger than I was, probably no more than 4 or 5 - and how I would make him laugh by using swear words. The thing was, he knew they were swear words, which is why he would laugh at them, and while I kind of knew that they were classed as such, it would not be until some time later - maybe years, I'm not sure - that I would be able to "feel" that they were and recognize them as such.
So knowing that event happened when I was 7 allows me to date the first memory of my existence at least a year or two before that.
Paul Robinson <postmaster@paul.washington.dc.us>
http://paul.washington.dc.us
Among other things, "A Philosopher, Computer Programmer, and Notary Public in and for the Commonwealth of Virginia."
"Above all else... We shall go on..."
"And continue!"
If this is supposed to produce a million customers over a three-month period, either it produces something in the neighborhood of $10 million a month if it charges out at $10 a month, or it produces in the neighborhood of $10 million a year if it's $10 a year. All up-front cash since the customer pays when he signs up. For that kind of money, either build the place to support the load or close down, refund the customer's money and stop complaining. You don't underfund the necessary equipment to run an operation then complain how you can't run it successfully on less than was necessary to operate it in the first place. That's whining and that's exactly what he's doing. Do it right, make it right, or admit you can't and close up.
Well, fine, we're talking about a business that should conceivably return several million a year in profit after deducting costs. That means that they should expect to put in the investment necessary to support the operation, or don't expect to get that big that fast and don't spend as much. Do one or the other. Anything else is dishonest. At the time I did not know the sizes involved. So raise the cost estimates I give in my other example by a factor of 20, and you get $60,000 for disk space for three years. And realize you don't need it all at once, when you drop below 50% free you buy another bunch of boxes or whatever. But I believe my original point is correct. Besides that, if it's that large an operation, either it should be spun off as a separate company where that is the only thing they do, or it should probably be operated on a managed hosting system which can probably handle the care and feeding of servers better than they can. Do something correctly or find someone who can, don't do a half-way job and then complain about how expensive it is. I guess it is this sort of dishonest way of operating projects is the reason more than 1/2 of all software projects get cancelled. Or is it 3/4? Improperly funding a project and incompetent management are the usual causes of that sort of failure. I've never worked in a big business because whenever I've applied at one of those places, the usual comment was I was overqualified for the position. There were a number of times I wanted to paste someone in the mouth for insulting me that way. To me, saying you won't hire someone because they are too good is saying that they want nothing but incompetents and are afraid of people who know what they are doing. Maybe that's the reason the manager of this system is whining instead of doing it right in the first place.Paul Robinson <postmaster@paul.washington.dc.us>
In my example I proposed the idea of running dual mirrored machines as RAID boxes. Are you trying to tell me that two identical machines running on an operating system that, according to the published reports, provides uptime reliability measured in years between reboots other than for planned outages is inadequately reliable for most transactional based systems? Then make it 4 machines, or 6 or 8 or whatever is reasonable.
There are two ways to increase reliability. One is to harden the system to prevent failure by building expensive boxes of extremely high reliability, with redundant components, dual processors, dual power supplies, etc. The other is to reduce the complexity of the system so that any particular failure at any point does not compromise the system.
Maybe I am wrong, but let's consider the cost of setting up a system consisting of 16 cheap computers each on a separate UPS, each containing a single inexpensive 60 billion byte drive and a gigabit network card, so that the system represents 4 of each drive group mirroring all the data sent to that group, doesn't that then provide the equivalent of 4 x 60 billion bytes, or 480 GB of storage, with at least four times the reliability (since you now have to have 4 boxes all simultaneously fail before any one "drive" is out of service, and each one only has to respond to 25% of the read requests for that group) at a fraction of the cost of a more expensive system that it in and of itself supposedly more reliable? Further, I think it's even more reliable since you can conceivably continue to operate, even in degraded mode, until 15 of the boxes fail which I do believe the possibility of failure only grows linearly while reliability grows exponentially as more machines are added.
A specific example of this is in the use of birth control. Any of the usual methods, pills, barriers, or chemicals, might have a 1-3% failure rate, i.e. 3 in 100. So, if a woman uses two forms of contraception, say diaphragm and foam together, the worst failure rate goes from 3 in 100 (3%) to 9 in 10000 (0.009%). If she can talk the guy into also using a condom, the failure rate is now 27 in 1,000,000 (0.000027%). Increase the number of things that have to fail to cause catastrophe and the reliability rate goes up.
I was giving a gross example where you can do the job with inexpensive equipment. So I could be wrong on some of the figures. So maybe it's a bit more expensive. But I do believe you can do this sort of thing successfully with the kind of setup I was talking about for less than $5,000 in any case. And remember, what I talking about here was moving the disk space to inexpensive mirrored servers, I said nothing about the transaction server(s).
So if you wouldn't recommend Linux (or BSD or any of the other similar items) what would you recommend? I would like to learn something here if I am wrong, please explain to me how this isn't going to work. I would like to know because I do want to understand and as I understand, this seems to be the most cost effective way to do this. I could be wrong and I'd like to understand more.
Below I will comment on how I was using the phrase "character data" in my article and the way I phrased it, I may have caused you to misunderstand me. I was under the impression that it does scale. If you feel that Linux is inadequate to handle the scaling factors for a large database system such as a game would use, tell me what you think such an application requires. Considering that I've heard that large clustering systems of thousands of nodes use Linux successfully to handle the load, and IBM has it on S/390 mainframes which are designed to handle huge transaction volumes and conceivably run thousands of Linux servers on one machine, I believed it did scale quite beautifully. But tell me so I can learn something if you think I am wrong. My apologies. When I said "character data" I meant the data to store the values of a character in a game, not character data as in text. I realize that the way I said it could have mislead you and I apologize for the way I said it as I realize now how it could have been misunderstood.I've worked with databases where the base system, with no data, just empty tables and stored procedures, ran in excess of 100 megabytes. I am saying that I think a factor of 10 megabytes average space per user is a lot of space. But maybe I'm wrong so let me see. If you figure a index key per record at being perhaps 16 bytes, and maybe you have 500 tables, and each table perhaps uses an average of 3, no let's say 5 links per record, and lets say the user has, oh, say, 50 records each in the system, that's 500x50x5x16 means you need about 2 megabytes for the indexes for each user. If you use normalized databases you might use maybe 2K per record based on trying not to repeat data except where needed for performance, so you need 5000K or 5 meg for the user's data. Add into that perhaps 3 meg for image data, and you have the 10 megabytes I estimated as a lot of space.
Okay, then, I was wrong on it being a lot of space. Make it 20 meg average per user and cut my estimates in half or double the cost.
If I am wrong on some of these numbers, please enlighten me so that when I make a statement in the future I will have the correct numbers and be able to justify my estimates as I am trying to do now, using my best estimates from my own experience in programming and database management. I know I do not know everything. But I can make educated guesses and perhaps, if someone else knows more than I do they can give me information to allow me to learn where I have made assumptions not backed up by real-life data.
And I wasn't figuring that to be a lifetime cost, I was figuring that based on the estimated 3 year life-span that an IDE disk is expected to have. If you have 48,000 customers each paying $10 a month to use the service - which is what I think many of these are doing - that's $480,000 a month, which is a lot of money. Based on usual cost figures of 50% for labor cost, this gives you $240,000 a month for salaries, $24,000 a month for G&A, (general and administrative ovehead) and perhaps another $48,000 a month for maintenance and equipment, and maybe another $12,000 a month for the Internet connections over multiple . This leaves in the neighborhood of $324,000 in costs which may be high. That means there is, or should be, something in the neighborhood of $100,000 a month to pay back the investors. As there are more people taking the service, it produces more income.
If the service is only being paid $10 a year, then you can run it through automated sign ups and buy managed hosting, and increase the amount of hosting as more people use the service.
The point I am making is that if you want to run a successful business you run it as an investment. You put some money in as initial capital, in the amounts necessary to get it started, according to what is reasonable to make it operational, and you expect it to eventually return cash in order to build up to larger capacities, and if it can't, you either make it pay or you close it. You do not complain that the costs of running the business make it non-economic. Then you're just whining, and as far as I can see that's exactly what he is doing, whining about how much it costs to run the operation instead of working with what you can afford and building it up as it produces an income stream.
To me it sounds suspiciously like it was expected they could just throw some money into this - probably less than it should have had - and expect it to have immediate huge cashflow and huge returns, and when it didn't they were disappointed. It don't work that way. A successful investment means you start it small, and allow the business to grow itself.
I think someone once said that those who become emotional over their investments soon end up losing money over them. And his sort of whine sounds suspiciously like emotionalism.
Paul Robinson <postmaster@paul.washington.dc.us>
I am reminded of the TV show, The Flintstones, when Fred has gotten a part in a movie playing the creature from the tar pits, and they're supposed to have safety equipment so he isn't hurt. The director has a scene where the hero uses a club on the creature, and wants a padded club for the scene. The property manager whines, "Are you kidding? Do you know how much padded clubs cost?" So they'll use a real one. You can guess the result: Fred is knocked cold on the first swing to the head. It's cheaper to injure extras than protect them.
If it's that expensive they perhaps should run their own server, either renting it or collocate a box of their own and put in enough disk space. If a T1 for network traffic costs about $500 a month, disk space is peanuts in comparison.
Last time I checked, a 120 GB - that's 120,000 megabytes - IDE disk sells for about $300. Buy an inexpensive used computer, say a 200 MHZ pentium for about $50 with a $75 Gigabit network card connected to the server via a $20 crossover cable with the same $75 gigabit network card so you can access disk space fast. So if you have a box with 4 of the 120 GB space, you can run 480 GB of space for all the characters - that's 480,000 megabytes - for $1500. Or you can mirror it and have 280GB of space completely mirrored. Or spend perhaps another $500, and put two drives each in separate machines with a gigabit switch between them. Or maybe the drives aren't that fast and you can get by with $15.00 100 megabit networking cards and a $20 hub. (You put the traffic going to and from the external disk drives on a separate network card so the data traffic doesn't interfere with the network traffic which is playing the game.)
If it takes as much as 10 megabytes per user (and that's a hell of a lot of space to store character data, and probably isn't anywhere near that much), you can host 48,000 customers with mirrored disk drives for a one-time cost of about $2K. And I believe all the software to handle this including raid striping or mirroring is built in to Linux which means it's free. If your provider is charging too much for disk space, colocate and reduce your cost. Or charge what it costs to operate the game.
This is just another way that companies that operate on-line games get greedy because they don't want the customers figuring ways to make money off the game (by selling characters they've built up) or want to impose draconian conditions (like no fan sites or no creating mods to their game) because they think the customer and any money he spends related to this product (and any attention he devotes to it) is all theirs and nobody else should be able to make any money at all related to it unless they get it or a huge percentage, or any non-company attention unless they control how that attention is given.
Anyone notice how when movies are released to theatres, the studio gets 90% of the ticket price? (The only thing keeping movie theatres alive is the popcorn stand.) Also notice how many studios, when they release films or tv shows where a star in it took a smaller salary in exchange for profit participation, discovers after millions of dollars in sales, that the show or movie has always lost money and thus allegedly never had any profit to pay out? I think it's no coincidence that some of these on-line games are run by divisions of studios and they are operating them the same way as the studios are run. They think everything related to their properties should be all theirs and resent anyone getting anything out of it unless they get a piece of the action. No, make that, they get as much as possible of the action, and resent sharing anything they might have to allow.
Paul Robinson <postmaster@paul.washington.dc.us>