Now, even assuming there's something remaining after thermite, how do you get it out of a molten platter? The head hovers at nanometers from the disk's surface. A bent disk with a huge hole through it will just instantly wreck any head trying to read it. Is it even technically possible to restore the platter to a condition where you can even try to read anything from it?
Besides, shouldn't all the data vanish due to the reaction bringing the surface above the Curie temperature?
Well, see, by that logic, DDOSing a server is completely fine. After all, a web server imposes a series of rules on you as well, such as what file you can request. If the rules it uses let you take it down, why would it be wrong to do that?
BTW, Second Life does have actual property and items that cost actual cash
If you need a huge multinational to make your work practical, then chances are you work at one already, which means your patents don't belong to you. If you're an individual, then it's possible this multinational will just choose to infringe on your patent and try to bankrupt you in court.
Unless you're really clever, they probably can make a minor tweak to your invention and release something very similar, just not covered by your patent anymore. May be cheaper for them to invest their own resources than to pay you.
Another thing is, how many companies will spend a large amount of cash on something that's not under their control? They'll want full rights to the patent, and after that, excellent chances that you're screwed.
First of all, what part of AC power generation and motors requires millions to demostrate? Toy motors are more than enough, and you can easily make a crappy AC motor by hand by winding wire yourself. It can be done by children in a school lab.
Second, if you lack the money to bring your product to market, what do you want the patent for? The point of a patent is to promote progress. If it's going to sit in a dusty drawer somewhere then it goes against the whole idea. And if you do have the money to develop a product, then you can build a prototype.
In the beginning, it's trivial to keep writing thousands of lines per month without a problem. I easily do. Now, a product that's almost done is completely different. You don't have large missing features anymore, most of the design is done. What's left is tweaking and polishing.
There's where it slows down. Rather than "writing function foo that takes X and Y parameters and returns bar", the problem becomes "How do I make this change in an elegant manner, and how do I make sure it doesn't break? How would Bob's code react if I made this change? Should this be done by tweaking X, Y, or rewriting this part because this requirement wasn't taken into account when it was written?".
It's a whole lot more thinking and much less coding at that stage. When maintaining a program, it's quite frequent that several minutes to several hours go to trying to reproduce the bug, then perhaps an hour to try to figure out what to do about it, and then the fix often turns out to be 5 - 20 lines of code.
Unless the air is really humid where you are, it'll work a lot better than a normal fan, and for pretty much the same price. If you have the luck of living in a dry climate, then it'll work really well.
Yup, I know about cascaded updates. That doesn't make it being stop being a pain in the arse though, as it doesn't always work, and physical constraints get in the way. Now that you have a code that has some actual meaning for somebody, changing it becomes a pain. Disks are slow, searching millions of rows takes time, and unless you're lucky enough to use a smart database like postgresql you end up with tons of locks that grind things to a halt. Major pain if it happens often.
Take an example from Unix, for example. You have an username, but what's stored on the disk is a number. If you find you want your account renamed, all the admin has to touch one line in/etc/passwd instead of searching every file on disk and correcting the username stored in it.
IMO, serial numbers are nearly always the best identifier for almost everything. Say, clients, products, etc. Why would you want to force your users to come up with a primary key? If they need "Foobar Enterprises", then just provide a good search dialog and let them find that. Otherwise you end up with things like "FOOBARE" for a primary key, people get mixed up when they enter "FOOBARI" instead, etc.
That's the best case, usually you'll end up with the database full of primary keys like 'ASDF' and 'KITTENS' because somebody didn't want to come up with something that made sense that day or they got annoyed when their choice was already taken and just typed something random to get it to work and be done with it. And if it doesn't make sense, why not just avoid it completely and let the database give it a number?
Personally, I only make exceptions where there's a natural primary key.
Your second example has maintenance problems, btw. It happens that people make an invoice for the wrong company. Sometimes because they messed up, sometimes because the company has multiple accounts for whatever reason. The first way it's trivial to move an invoice, the second it isn't.
That leads to maintenance nightmares. I'm actually going in the reverse direction.
Integer keys are wonderful. Product #5 is just that, product #5. If you want to find something about it, join to the products table.
Sure, you can make a text product code, and have 'ENERGIZER_AA_BATTERY' instead of #5. That'll look good for a while. Until one day boss comes and tells you they messed up, and it actually should be 'AAA'. That's when you see the difference. With integer keys, the key is meaningless and eternal. You change 1 field in 1 table.
With the text key, however, you now need to fix all references to this product everywhere, which is easily 50 tables on a large DB. If you're unlucky and can't have the DB propagate changes correctly, you now need to make sure to update every single table, and not forget any.
But that's just the beginning, because obviously you have to do all this stuff inside a transaction. A transaction which in a database like MSSQL will create lots of locks that have excellent chances of bringing the whole DB to a screenching halt for whatever time it takes to do the update.
Sure, he deserved getting caught. But it's just disappointing that a guy that has enough brains to earn $1M that way didn't have enough to just vanish and live quietly in another country before the police came looking.
It's like the plot from a bank robbery movie! Thieves get in, steal traffic control equipment, then happily *eat and smoke* in the place, and drive away without hurry as the traffic lights lose sync. They even stole the alarm as well. There's something to be said about a job well done;-)
I tried with "compiler" and "gcc", and this is what I got:
Compiler vanishes past the transmitted skill. The imperative walks Compiler across the wrecker. Compiler milks GCC. The scarce symptom reverts throughout GCC. Compiler breaches a coin behind the uncertain knight.
GCC undertakes Compiler. Why does GCC base the amazed supplier? Compiler blames GCC under the psychologist. Compiler bangs GCC against the mercury. The producer strikes Compiler. GCC fingers Compiler.
My aunt is an university teacher in Russia. Last time I asked, her wage was somewhere about $100 a month. From what I hear, in America it's quite possible to earn $4000 a month as a programmer. Let's try a very conservative $2000 then. Let's say the CD costs $15 for you.
Now the question: Would anybody in their right mind buy the CD for $300, which is the russian equivalent? If you don't believe me, see "Living Wage in Russia not Enough for British Cat": http://english.pravda.ru/society/family/09-01-2004 /4540-cat-0
So, if music is going to be sold at all in Russia it's going to have to be at a reasonable price -- which is ridiculously small by American standards. AllOfMP3 AFAIK, pays royalties. Of course they're going to be very small by American standards as well. But this is globalization. You can't have your cake and eat it too.
While I agree that search engines are a pain with this kind of thing, Perl has several very nice manpages. Specifically, the "perlvar" one contains the answer to your question, and the "perl" manpage lists all the ones there are. From the manpage:
The index of the first element in an array, and of the first character in a substring. Default is 0, but you could theoretically set it to 1 to make Perl behave more like awk (or Fortran) when subscripting and when evaluating the index() and substr() functions. (Mnemonic: [ begins subscripts.)
As of release 5 of Perl, assignment to $[ is treated as a compiler directive, and cannot influence the behavior of any other file. (That's why you can only assign compile-time constants to it.) Its use is highly discouraged.
Note that, unlike other compile-time directives (such as strict), assignment to $[ can be seen from outer lexical scopes in the same file. However, you can use local() on it to strictly bind its value to a lexical block.
I could use something that doesn't need batteries, charges itself, and has a bright light that doesn't have to last very long. Such a thing would come very handy for messing with computers and such. It's quite a pain that when you just need a bit of extra light for a minute that the batteries have to be always dead.
Sooo, what your saying is selling to Furries is the number 1 SL business plan?
No, I'd say the number one business plan is becoming a land baron. But Anshe Chung has that covered already (and even employs several people for it!). Casinos and such seem to be very popular as well.
Making furry avatars probably can be done for good profit though. But you'd need to be damn good at it, Luskwood would be very hard to beat.
Why no consumers? Coding in SL isn't a very easy thing. To start with, LSL is their own language, which means you need to invest an effort to learn it. IMO, coding in SL isn't much easier than an actual coding job. Then you to have need business sense, advertise, etc.
There certainly seem to be quite a few consumers in SL. For example, making avatars doesn't require any special tools, but takes serious skill to get right. Any moron can attach a box to their head, but it takes months to develop the skill to make this: http://www.luskwood.com/. One of these costs about $3 US, which would make it worth it when the alternative is spending months to learn.
Of course, this can fail. It is possible to have too many producers indeed. But to suggest that it's some kind of pyramid scheme is nonsense, IMO. If SL gets flooded with people trying to make a living, some will be worthless as artists, and some bad at coding. With some luck, enough normal people will join as well. The first two will team up to sell to the later.
What's the natural resource involved when people write code?
From my limited experience, programming business in SL works as follows. You spend some time writing a LSL script. If that does something useful somebody else wants, another person pays $ to convert them to L$, and pays you in exchange for the scripted object. Then you convert L$ to $ and get real money.
Effectively, SL is acting as an easy and convenient way of connecting programmer with customer. No magic involved.
You read from a RAID, of course. If you have something like LTO3, which costs about $2.5K per drive (IIRC) you almost certainly have one of those already.
I think the point here was that tape *can* be faster than hard disks. To get a sustained 100MB/s you need at least a couple good drives, not the cheap consumer stuff.
IIRC, the better tape drives like LTO have the ability to slow down if data isn't coming fast enough. The cheaper ones like DDS will just stop the tape and sit there, which wears the tape, but DDS4 writes at about 2.5MB/s anyway.
Just send spam that's so obvious that every filter in existence will classify it as spam, and you'll have a nice botnet doing your bidding.
This, IMO, is the main problem here. This way of doing things certainly works. However, it requires coordination to avoid situations like above, and there have to be humans somewhere taking decisions.
The problem with that is: How do you organize a system where a bunch of humans vote on what is spam and what isn't, and then communicate this result in a secure and distributed way to the clients? And how do you avoid making it vulnerable to attempts to disrupt it?
Well, it's certainly possible. The trick is making it easy and reliable.
Say, a basic solution could work as follows: Use spamassassin to filter spam, take the stuff with the worst score, extract an URL, and wget say, up to 1MB from there.
The problem with that is that if everybody runs it this way, eventually somebody innocent gets DDOSed, when somebody in Marketing manages to write a mail that scores 20 points on spamassassin.
So, BlueSecurity came in by personally checking each spam to make sure (as I understand it). A replacement would need some way of making sure that only spammers are getting hammered, and that's difficult with a distributed system.
I've been itching to sign up since I heard of this here, but first it was no confirmation email, then the members site went for a whole week with a "we're reorganizing it" message. I was wondering what kind of moron they have as an admin.
This is extremely disappointing, I must say. Now that they finally got a noticeable success, world wide recognition and made lots of spammers squirm and wonder what will they do, they go and give up? Sheesh.
But ah well. The client was Open Source, wasn't it? So, who will pick this one up, and get it back running? Pretty much all of the work seems to be done already, all it seems to need is becoming distributed, which would avoid this situation in the future.
I've considered doing just this a few times. Every time I've tried, the various BSD install CDs have all failed hilariously in their attempts to boot my machine.
Have you considered that licensing might be an issue here? I have zero wish to work on BSD. Why would I let my work be used for free by some multinational? And why would any company give out their improvement under the BSD license if they're perfectly fine not to?
Let's take your case for instance. Your machine fails to boot. This could be fixed by either somebody who works on BSD for fun, or by some company who wants it to boot on some kind of machine. Now, if say, IBM, fixes this for Linux they *have* to give out the modification. They could keep it in-house, but that's not very useful.
On the other hand, probably some company already fixed that issue when say, making firmware for some embedded device. But why would they bother to give it back? It means extra time, and they get absolutely nothing out of it.
But anyway, my main point was the last bit (the "non-basement-dwelling" bit), that rules are ultimately defined by their enforcement, not by badly written emails sent by concerned nobodies.
Well, while I agree that the guy doesn't seem to have an actual case here, I don't agree with the point you're trying to make. Sure, he's got no real authority. He would be if he was a copyright holder. But these "nobodies" are perfectly capable of having an effect by alerting the EFF, or the BSA. I'm pretty sure the EFF would gladly get into this if there was a case.
While I completely agree that the guy is a moron, the point of the GPL certainly was never to "enable people to stop worrying about this licensing crap". You should really go and read what Stallman writes some time. The intention is certainly to make things open and free, and the license is a way of insuring that.
If you don't want the "licensing crap", I'd recommend switching to BSD.
What are you saying? That Linux was put together for the convenience of the Linux developers, but not for users?
Yes, exactly that! What, you just noticed? Linus made the kernel for *fun*. Of course it was put together for the convenience of developers! Where have you seen Linus claim he was trying to make a product to put on the shelf?
Slashdot Mirror
Now, even assuming there's something remaining after thermite, how do you get it out of a molten platter? The head hovers at nanometers from the disk's surface. A bent disk with a huge hole through it will just instantly wreck any head trying to read it. Is it even technically possible to restore the platter to a condition where you can even try to read anything from it?
Besides, shouldn't all the data vanish due to the reaction bringing the surface above the Curie temperature?
Well, see, by that logic, DDOSing a server is completely fine. After all, a web server imposes a series of rules on you as well, such as what file you can request. If the rules it uses let you take it down, why would it be wrong to do that?
BTW, Second Life does have actual property and items that cost actual cash
That doesn't work so well these days though.
If you need a huge multinational to make your work practical, then chances are you work at one already, which means your patents don't belong to you. If you're an individual, then it's possible this multinational will just choose to infringe on your patent and try to bankrupt you in court.
Unless you're really clever, they probably can make a minor tweak to your invention and release something very similar, just not covered by your patent anymore. May be cheaper for them to invest their own resources than to pay you.
Another thing is, how many companies will spend a large amount of cash on something that's not under their control? They'll want full rights to the patent, and after that, excellent chances that you're screwed.
Uh, no.
First of all, what part of AC power generation and motors requires millions to demostrate? Toy motors are more than enough, and you can easily make a crappy AC motor by hand by winding wire yourself. It can be done by children in a school lab.
Second, if you lack the money to bring your product to market, what do you want the patent for? The point of a patent is to promote progress. If it's going to sit in a dusty drawer somewhere then it goes against the whole idea. And if you do have the money to develop a product, then you can build a prototype.
It varies a lot though.
In the beginning, it's trivial to keep writing thousands of lines per month without a problem. I easily do. Now, a product that's almost done is completely different. You don't have large missing features anymore, most of the design is done. What's left is tweaking and polishing.
There's where it slows down. Rather than "writing function foo that takes X and Y parameters and returns bar", the problem becomes "How do I make this change in an elegant manner, and how do I make sure it doesn't break? How would Bob's code react if I made this change? Should this be done by tweaking X, Y, or rewriting this part because this requirement wasn't taken into account when it was written?".
It's a whole lot more thinking and much less coding at that stage. When maintaining a program, it's quite frequent that several minutes to several hours go to trying to reproduce the bug, then perhaps an hour to try to figure out what to do about it, and then the fix often turns out to be 5 - 20 lines of code.
Unless the air is really humid where you are, it'll work a lot better than a normal fan, and for pretty much the same price. If you have the luck of living in a dry climate, then it'll work really well.
Yup, I know about cascaded updates. That doesn't make it being stop being a pain in the arse though, as it doesn't always work, and physical constraints get in the way. Now that you have a code that has some actual meaning for somebody, changing it becomes a pain. Disks are slow, searching millions of rows takes time, and unless you're lucky enough to use a smart database like postgresql you end up with tons of locks that grind things to a halt. Major pain if it happens often.
/etc/passwd instead of searching every file on disk and correcting the username stored in it.
Take an example from Unix, for example. You have an username, but what's stored on the disk is a number. If you find you want your account renamed, all the admin has to touch one line in
IMO, serial numbers are nearly always the best identifier for almost everything. Say, clients, products, etc. Why would you want to force your users to come up with a primary key? If they need "Foobar Enterprises", then just provide a good search dialog and let them find that. Otherwise you end up with things like "FOOBARE" for a primary key, people get mixed up when they enter "FOOBARI" instead, etc.
That's the best case, usually you'll end up with the database full of primary keys like 'ASDF' and 'KITTENS' because somebody didn't want to come up with something that made sense that day or they got annoyed when their choice was already taken and just typed something random to get it to work and be done with it. And if it doesn't make sense, why not just avoid it completely and let the database give it a number?
Personally, I only make exceptions where there's a natural primary key.
Your second example has maintenance problems, btw. It happens that people make an invoice for the wrong company. Sometimes because they messed up, sometimes because the company has multiple accounts for whatever reason. The first way it's trivial to move an invoice, the second it isn't.
Uhh... nanotubes easy to break? Since when? They're the only material strong enough for a practical space elevator! Think a bit about that.
That leads to maintenance nightmares. I'm actually going in the reverse direction.
Integer keys are wonderful. Product #5 is just that, product #5. If you want to find something about it, join to the products table.
Sure, you can make a text product code, and have 'ENERGIZER_AA_BATTERY' instead of #5. That'll look good for a while. Until one day boss comes and tells you they messed up, and it actually should be 'AAA'. That's when you see the difference. With integer keys, the key is meaningless and eternal. You change 1 field in 1 table.
With the text key, however, you now need to fix all references to this product everywhere, which is easily 50 tables on a large DB. If you're unlucky and can't have the DB propagate changes correctly, you now need to make sure to update every single table, and not forget any.
But that's just the beginning, because obviously you have to do all this stuff inside a transaction. A transaction which in a database like MSSQL will create lots of locks that have excellent chances of bringing the whole DB to a screenching halt for whatever time it takes to do the update.
Sure, he deserved getting caught. But it's just disappointing that a guy that has enough brains to earn $1M that way didn't have enough to just vanish and live quietly in another country before the police came looking.
;-)
It's quite amazing what people manage to pull off sometimes, like this: http://www.wired.com/news/business/1,52114-0.html
It's like the plot from a bank robbery movie! Thieves get in, steal traffic control equipment, then happily *eat and smoke* in the place, and drive away without hurry as the traffic lights lose sync. They even stole the alarm as well. There's something to be said about a job well done
Very funny, but it has a long way to go.
No, it's completely right.
4 /4540-cat-0
My aunt is an university teacher in Russia. Last time I asked, her wage was somewhere about $100 a month. From what I hear, in America it's quite possible to earn $4000 a month as a programmer. Let's try a very conservative $2000 then. Let's say the CD costs $15 for you.
Now the question: Would anybody in their right mind buy the CD for $300, which is the russian equivalent? If you don't believe me, see "Living Wage in Russia not Enough for British Cat": http://english.pravda.ru/society/family/09-01-200
So, if music is going to be sold at all in Russia it's going to have to be at a reasonable price -- which is ridiculously small by American standards. AllOfMP3 AFAIK, pays royalties. Of course they're going to be very small by American standards as well. But this is globalization. You can't have your cake and eat it too.
Yes, the availability of the source for Quake 1 and 2 sure killed Quake 3 sales, didn't it?
I could use something that doesn't need batteries, charges itself, and has a bright light that doesn't have to last very long. Such a thing would come very handy for messing with computers and such. It's quite a pain that when you just need a bit of extra light for a minute that the batteries have to be always dead.
Sooo, what your saying is selling to Furries is the number 1 SL business plan?
No, I'd say the number one business plan is becoming a land baron. But Anshe Chung has that covered already (and even employs several people for it!). Casinos and such seem to be very popular as well.
Making furry avatars probably can be done for good profit though. But you'd need to be damn good at it, Luskwood would be very hard to beat.
Why no consumers? Coding in SL isn't a very easy thing. To start with, LSL is their own language, which means you need to invest an effort to learn it. IMO, coding in SL isn't much easier than an actual coding job. Then you to have need business sense, advertise, etc.
There certainly seem to be quite a few consumers in SL. For example, making avatars doesn't require any special tools, but takes serious skill to get right. Any moron can attach a box to their head, but it takes months to develop the skill to make this: http://www.luskwood.com/. One of these costs about $3 US, which would make it worth it when the alternative is spending months to learn.
Of course, this can fail. It is possible to have too many producers indeed. But to suggest that it's some kind of pyramid scheme is nonsense, IMO. If SL gets flooded with people trying to make a living, some will be worthless as artists, and some bad at coding. With some luck, enough normal people will join as well. The first two will team up to sell to the later.
So how do programmers make a living in RL, then?
What's the natural resource involved when people write code?
From my limited experience, programming business in SL works as follows. You spend some time writing a LSL script. If that does something useful somebody else wants, another person pays $ to convert them to L$, and pays you in exchange for the scripted object. Then you convert L$ to $ and get real money.
Effectively, SL is acting as an easy and convenient way of connecting programmer with customer. No magic involved.
You read from a RAID, of course. If you have something like LTO3, which costs about $2.5K per drive (IIRC) you almost certainly have one of those already.
I think the point here was that tape *can* be faster than hard disks. To get a sustained 100MB/s you need at least a couple good drives, not the cheap consumer stuff.
IIRC, the better tape drives like LTO have the ability to slow down if data isn't coming fast enough. The cheaper ones like DDS will just stop the tape and sit there, which wears the tape, but DDS4 writes at about 2.5MB/s anyway.
Forgot to add that this would work right until the point the spammers start sending mail like the following:
"Buy viagra xanax prozac at http://microsoft.com/"
Just send spam that's so obvious that every filter in existence will classify it as spam, and you'll have a nice botnet doing your bidding.
This, IMO, is the main problem here. This way of doing things certainly works. However, it requires coordination to avoid situations like above, and there have to be humans somewhere taking decisions.
The problem with that is: How do you organize a system where a bunch of humans vote on what is spam and what isn't, and then communicate this result in a secure and distributed way to the clients? And how do you avoid making it vulnerable to attempts to disrupt it?
Well, it's certainly possible. The trick is making it easy and reliable.
Say, a basic solution could work as follows: Use spamassassin to filter spam, take the stuff with the worst score, extract an URL, and wget say, up to 1MB from there.
The problem with that is that if everybody runs it this way, eventually somebody innocent gets DDOSed, when somebody in Marketing manages to write a mail that scores 20 points on spamassassin.
So, BlueSecurity came in by personally checking each spam to make sure (as I understand it). A replacement would need some way of making sure that only spammers are getting hammered, and that's difficult with a distributed system.
I've been itching to sign up since I heard of this here, but first it was no confirmation email, then the members site went for a whole week with a "we're reorganizing it" message. I was wondering what kind of moron they have as an admin.
This is extremely disappointing, I must say. Now that they finally got a noticeable success, world wide recognition and made lots of spammers squirm and wonder what will they do, they go and give up? Sheesh.
But ah well. The client was Open Source, wasn't it? So, who will pick this one up, and get it back running? Pretty much all of the work seems to be done already, all it seems to need is becoming distributed, which would avoid this situation in the future.
I've considered doing just this a few times. Every time I've tried, the various BSD install CDs have all failed hilariously in their attempts to boot my machine.
Have you considered that licensing might be an issue here? I have zero wish to work on BSD. Why would I let my work be used for free by some multinational? And why would any company give out their improvement under the BSD license if they're perfectly fine not to?
Let's take your case for instance. Your machine fails to boot. This could be fixed by either somebody who works on BSD for fun, or by some company who wants it to boot on some kind of machine. Now, if say, IBM, fixes this for Linux they *have* to give out the modification. They could keep it in-house, but that's not very useful.
On the other hand, probably some company already fixed that issue when say, making firmware for some embedded device. But why would they bother to give it back? It means extra time, and they get absolutely nothing out of it.
But anyway, my main point was the last bit (the "non-basement-dwelling" bit), that rules are ultimately defined by their enforcement, not by badly written emails sent by concerned nobodies.
Well, while I agree that the guy doesn't seem to have an actual case here, I don't agree with the point you're trying to make. Sure, he's got no real authority. He would be if he was a copyright holder. But these "nobodies" are perfectly capable of having an effect by alerting the EFF, or the BSA. I'm pretty sure the EFF would gladly get into this if there was a case.
While I completely agree that the guy is a moron, the point of the GPL certainly was never to "enable people to stop worrying about this licensing crap". You should really go and read what Stallman writes some time. The intention is certainly to make things open and free, and the license is a way of insuring that.
If you don't want the "licensing crap", I'd recommend switching to BSD.
What are you saying? That Linux was put together for the convenience of the Linux developers, but not for users?
Yes, exactly that! What, you just noticed? Linus made the kernel for *fun*. Of course it was put together for the convenience of developers! Where have you seen Linus claim he was trying to make a product to put on the shelf?
So some people won't use it. Big deal.