I beg to differ here: each and every piece of software must by definition, set a 'machine state'.
But not a specific machine state. As I mentioned before, the choice of compiler will affect the machine state even when the result of the program is unchanged; the processor family (Intel to Alpha) will involve a totally different machine state for the same high-level code.
an explicit and specific arrangement of gates
This is only true in the cases where the program is for a gate array or some other specialised system and even then there are normally more than one configuration that will produced the desired result for anything more than the most trivial of programs.
equivalent to a chemical process, or really, any manufacturing process wherein resource A is combined or manipulated to produce resource B.
This is partly true but it has been traditional to distinguish information production from material production. For example, a film transforms the entropy in a white light source into a pattern on a screen - producing information for the viewer. Again, copyright is the normal method of preventing plagarism in these fields, not patents. One reason for this is that films, stories, and computer programs can have their processes changed (Film of the play of the book; recompilation of the program) but still remain the same thing. This is unlike a chemical process which is requred under patent law to be quite specific in how it is performed.
but the only thing mechanical engineers have at the end of the day are machines that can be patented.
But we're not talking about mechanical engineers, we're talking about logical engineers who are building not machines but data tranformations which will be carried out by machines built by someone else for the express purpose of performing such tranformations (or "programs").
Why should I be able to patent a method of using your machine when the whole point of your machine it allow it to be so used? This is like patenting a novel because it uses paper in a unique combination of ink-markings to produce a "novel invention". Copyright is much closer to the correct solution than patent law.
Computer Code clearly falls under the auspices of Patent Law, and nothing further. It is a purely mechanical system, and the code is ultimately just shorthand to arrive at the desired effect. In other words, a diagram. Just as a lawnmower or lightbulb would require for a patent. That it is inconvenient to show a physical diagram of software is irrelevant, just as it is irrelevant to copyright computer code.
Patent law has nothing to do with software. The mechanical argument fails as the intent of a program is not the setting of gates (so a program rarely deals with the gates), it is the production of a desired transformation of some input data to some output data. This can normally be achieved in many different ways. Indeed the only examples that can't be done in different ways would very well serve as the definition of "obvious" when attepting to overturn the patent.
Patent law does not cover effects and two patents can be issued to two different lightbulbs which achieve their effects in two different ways.
Since the method of "setting the gates" for a software effect varies, even with the same source code, based on the processor, memory configuration, and compiler version it is impractical to the point of impossibility to apply the mechanical argument. There is simply no possibility of enforcing such an approach in software and there isn't even very much logic to trying.
The strongest approach is copyright on the source code itself which abstracts the whole gates thing to a level where it is at least possible to debate whether something has been copied or not.
Even this, however, is difficult as differing programming languages have some very different ways of expressing the same thing. Try APL, Forth, Lisp and C++ for multiplying two matricies; could you prove that they had been copied from each other (or that they had not)?
The final approach is to patent/copyright an algorithm. This, though, is a real can of worms. Since an algorithm is simply a list of instructions, allowing the protection of these as if they were property means that there is no reason I can't "own" the best route from my street to the shops and route-finder programs would be a legal nightmare. Even cookery books would become rich pickings for lawyers as these are simply collections of algorithms (if we split algorithms for computers into a special category and ignore everything else what happens when a machine is programmed to make bread, is that covered or not?).
So, perhaps the "(self-styled) brightest minds" of the age have actually thought about the difficulties involved in these issues after all. Isn't that reassuring?
But for 99% of the text you see, it doesn't do much.
When using fonts not specifically designed for low-res (300dpi and under) it does make a big difference; I find the web really clear in GillSans but without AA Gill just turns into blobs in many situations.
I read the conditions of use page twice and couldn't see any such claim. Admittedly it was so boring I might have dozed off and missed it. Have you seen this claim on the actual site?
Of course they keep a copy of every paper, that's how they check for plagarism! What did you think they did use the md5 hash cross-indexed with a Tarot reading?
Pre-built components, such as scrollbars and buttons, are included
Just like my web browser.
The sooner these bozos go bust the better; 90% of all badly designed websites use Flash, it's good for nothing at all.
TWW
Re: It's my trumpet and I'll blow it if I want to
on
Apple Licenses CUPS
·
· Score: 1
The issue I had was under LPD that trying to associate the same device (/dev/lp) with different queues (colour and draft, for example) would result in confusion and timeouts for one or other of the queues as the lpd sub-daemon looking after the queue saw the device as being unavailable.
TWW
Re: It's my trumpet and I'll blow it if I want to
on
Apple Licenses CUPS
·
· Score: 1
Or just use lpd.
Lpd has a lot of problems if you want to set up the printer for easy access to all the various modes that these printers support.
TWW
It's my trumpet and I'll blow it if I want to
on
Apple Licenses CUPS
·
· Score: 4, Interesting
That's nice, but when can I print over the network to my Epson inkjet, like I can in Mac OS 9 with USB Printer Sharing?
You could try the alpha version of my own printing system which I've written in Perl after three attempts to get CUPS to work ended in failure. I use it to print across the network to my Epson S.P. 1290.
The problem is that there is competition out there
Is it really that tough in your sector? What are the figures as regards traffic to sites that geeks frequent; I find it hard to imagine that/. isn't one of the top three.
Either way, what are you going to do if advertisers say they aren't going to play if you make a big push for replacing them with subscriptions?
I posted this earlier but I haven't heard from anyone that can explain why/. is thinking this way, so here it is again:
So what you're saying is "We've got 1/3 million users per day and we've got to do what the advertisers want"?
Well, Jesus, how many readers do you need before you start telling the advertisers what they have to do to get on?
If that really is the state of on-line marketing then you'd be better off getting out of it and selling blank discs on street corners because that situation is not stable.
What happens if the advertisers say "Dump the no-ads pages or we walk"?
If you have the traffic you say you have then you should be telling them to get into an orderly line and, while they're waiting they can read your advertisement restrictions. I repeat, if you can't do this then you will never have a stable business model that includes advertising.
Dumping the adverts altogether and making posts a subscription-only facility (read and write) would be better. Why arse around with people that want big adverts on the site for next to nothing in the first place? Tell them where to stick it.
Where do you draw the line on the last item? I can make my company nt-for-profit by giving everyone a pay rise when there's a profit and cutting their pay when there's not.
So what you're saying is "We've got 1/3 million users per day and we've got to do what the advertisers want"?
Well, Jesus, how many readers do you need before you start telling the advertisers what they have to do to get on?
If that really is the state of on-line marketing then you'd be better off getting out of it and selling blank discs on street corners because that situation is not stable.
What happens if the advertisers say "Dump the no-ads pages or we walk"?
CmdrTaco: if you let moderators do stupid things like this (and yes, there are more stupid moderators than smart ones), you're going to alienate all your clueful posters.
This is what meta-moderation is for, if someone disagrees with the moderation.
However, I wouldn't mark this particular moderator down. Your post was a list of complaints about what other people are doing to solve a real problem with no solution of your own to offer or any reason why its not a problem. Personally I wouldn't mark it as a troll but I think it's borderline.
In particular, Steve Deering, the author of SIP, which later merged with a number of other things to become IPv6, would agree that 128 bit IP addresses is overkill, and SIP had only 64 bit addresses originally.
See, this is what you should have put in the original post. On this point, is there any reason not to go to 128bits now and save ever having to do this again, even when there are toasters on Mars? Is 64bits really enough if nano-tech machines ever need to be addressed and everyone's watch is on a broadband Internet connection? Probably, yes, but think of the hassle if it ever needs upgraded again. Better to sort it out now forever.
MS today announced that they would stop hiring programmers that used HTTP for everything. They denied that they had considered scrapping HTTP because their current programmers didn't know how to use it properly. "That would be insane; like banning paint because lots of people have poor taste in decorating their homes." an official said.
If this new method speeds the calculation by a factor of three, and each extra bit in the keys doubles the amount of time needed then surely this "breakthough" amounts to everyone losing less than 1.5 bits of security, doen't it?
The poster seems to think speeding the calculation by 3x means reducing the strength of 300bits to that of 100bits. I know this is plain wrong but I'm not sure of the correct value.
The issue is that, since 99.9999% of all DNS servers on the 'Net don't know who you or your new DNS server are the change has zero practical effect. ICANN exists only because it's a monopoly and can not be challanged at a technical level because of that monopoly. Legal recourse is also out since all politicians with real power are corrupt and/or technically incompetent.
Every vote for or against a politician is a vote for the system. ICANN is a perfect example of just how unimportant ethics or "the people" are in the face of $$$$'s.
I generally agree, but the category "ANIMATED FEATURE FILM" is something I think we can do without. Monsters Inc should be in the running for BEST FILM, not some "best weirdo minority thing". These sorts of categories just ghetto-ise minorities or turn them into self-reflective little universes that have no, and don't try to have, relevence outside themselves (I'm thinking MOBOs here, although at least ANIMATED FEATURE FILM isn't a racist award).
All the Pixar films so far should have at least been nominated for Best Film of their year, I think the time when that can happen is moved further away by this category.
Slashdot Mirror
Actually, I was only agreeing with the principle involved; I think that 5 years would probably be better too.
I'm not going to try to argue you out of it since I agree!
TWW
But not a specific machine state. As I mentioned before, the choice of compiler will affect the machine state even when the result of the program is unchanged; the processor family (Intel to Alpha) will involve a totally different machine state for the same high-level code.
an explicit and specific arrangement of gates
This is only true in the cases where the program is for a gate array or some other specialised system and even then there are normally more than one configuration that will produced the desired result for anything more than the most trivial of programs.
equivalent to a chemical process, or really, any manufacturing process wherein resource A is combined or manipulated to produce resource B.
This is partly true but it has been traditional to distinguish information production from material production. For example, a film transforms the entropy in a white light source into a pattern on a screen - producing information for the viewer. Again, copyright is the normal method of preventing plagarism in these fields, not patents. One reason for this is that films, stories, and computer programs can have their processes changed (Film of the play of the book; recompilation of the program) but still remain the same thing. This is unlike a chemical process which is requred under patent law to be quite specific in how it is performed.
What's the problem with that?
TWW
At least it's well known so it's easy to add to the spam blockers. Hope he didn't have anything he wanted to send me in an email.
TWW
But we're not talking about mechanical engineers, we're talking about logical engineers who are building not machines but data tranformations which will be carried out by machines built by someone else for the express purpose of performing such tranformations (or "programs").
Why should I be able to patent a method of using your machine when the whole point of your machine it allow it to be so used? This is like patenting a novel because it uses paper in a unique combination of ink-markings to produce a "novel invention". Copyright is much closer to the correct solution than patent law.
TWW
TWW
Patent law has nothing to do with software. The mechanical argument fails as the intent of a program is not the setting of gates (so a program rarely deals with the gates), it is the production of a desired transformation of some input data to some output data. This can normally be achieved in many different ways. Indeed the only examples that can't be done in different ways would very well serve as the definition of "obvious" when attepting to overturn the patent.
Patent law does not cover effects and two patents can be issued to two different lightbulbs which achieve their effects in two different ways.
Since the method of "setting the gates" for a software effect varies, even with the same source code, based on the processor, memory configuration, and compiler version it is impractical to the point of impossibility to apply the mechanical argument. There is simply no possibility of enforcing such an approach in software and there isn't even very much logic to trying.
The strongest approach is copyright on the source code itself which abstracts the whole gates thing to a level where it is at least possible to debate whether something has been copied or not.
Even this, however, is difficult as differing programming languages have some very different ways of expressing the same thing. Try APL, Forth, Lisp and C++ for multiplying two matricies; could you prove that they had been copied from each other (or that they had not)?
The final approach is to patent/copyright an algorithm. This, though, is a real can of worms. Since an algorithm is simply a list of instructions, allowing the protection of these as if they were property means that there is no reason I can't "own" the best route from my street to the shops and route-finder programs would be a legal nightmare. Even cookery books would become rich pickings for lawyers as these are simply collections of algorithms (if we split algorithms for computers into a special category and ignore everything else what happens when a machine is programmed to make bread, is that covered or not?).
So, perhaps the "(self-styled) brightest minds" of the age have actually thought about the difficulties involved in these issues after all. Isn't that reassuring?
TWW
What happens if someone's found an old copy on Google's cache or something?
TWW
When using fonts not specifically designed for low-res (300dpi and under) it does make a big difference; I find the web really clear in GillSans but without AA Gill just turns into blobs in many situations.
TWW
TWW
I read the conditions of use page twice and couldn't see any such claim. Admittedly it was so boring I might have dozed off and missed it. Have you seen this claim on the actual site?
TWW
TWW
Pre-built components, such as scrollbars and buttons, are included
Just like my web browser.
The sooner these bozos go bust the better; 90% of all badly designed websites use Flash, it's good for nothing at all.
TWW
TWW
Lpd has a lot of problems if you want to set up the printer for easy access to all the various modes that these printers support.
TWW
You could try the alpha version of my own printing system which I've written in Perl after three attempts to get CUPS to work ended in failure. I use it to print across the network to my Epson S.P. 1290.
TWW
Is it really that tough in your sector? What are the figures as regards traffic to sites that geeks frequent; I find it hard to imagine that /. isn't one of the top three.
Either way, what are you going to do if advertisers say they aren't going to play if you make a big push for replacing them with subscriptions?
TWW
I posted this earlier but I haven't heard from anyone that can explain why /. is thinking this way, so here it is again:
So what you're saying is "We've got 1/3 million users per day and we've got to do what the advertisers want"?
Well, Jesus, how many readers do you need before you start telling the advertisers what they have to do to get on?
If that really is the state of on-line marketing then you'd be better off getting out of it and selling blank discs on street corners because that situation is not stable. What happens if the advertisers say "Dump the no-ads pages or we walk"?
If you have the traffic you say you have then you should be telling them to get into an orderly line and, while they're waiting they can read your advertisement restrictions. I repeat, if you can't do this then you will never have a stable business model that includes advertising.
Dumping the adverts altogether and making posts a subscription-only facility (read and write) would be better. Why arse around with people that want big adverts on the site for next to nothing in the first place? Tell them where to stick it.
TWW
Where do you draw the line on the last item? I can make my company nt-for-profit by giving everyone a pay rise when there's a profit and cutting their pay when there's not.
TWW
So what you're saying is "We've got 1/3 million users per day and we've got to do what the advertisers want"?
Well, Jesus, how many readers do you need before you start telling the advertisers what they have to do to get on?
If that really is the state of on-line marketing then you'd be better off getting out of it and selling blank discs on street corners because that situation is not stable.
What happens if the advertisers say "Dump the no-ads pages or we walk"?
TWW
This is what meta-moderation is for, if someone disagrees with the moderation.
However, I wouldn't mark this particular moderator down. Your post was a list of complaints about what other people are doing to solve a real problem with no solution of your own to offer or any reason why its not a problem. Personally I wouldn't mark it as a troll but I think it's borderline.
In particular, Steve Deering, the author of SIP, which later merged with a number of other things to become IPv6, would agree that 128 bit IP addresses is overkill, and SIP had only 64 bit addresses originally.
See, this is what you should have put in the original post. On this point, is there any reason not to go to 128bits now and save ever having to do this again, even when there are toasters on Mars? Is 64bits really enough if nano-tech machines ever need to be addressed and everyone's watch is on a broadband Internet connection? Probably, yes, but think of the hassle if it ever needs upgraded again. Better to sort it out now forever.
TWW
TWW
The poster seems to think speeding the calculation by 3x means reducing the strength of 300bits to that of 100bits. I know this is plain wrong but I'm not sure of the correct value.
TWW
Every vote for or against a politician is a vote for the system. ICANN is a perfect example of just how unimportant ethics or "the people" are in the face of $$$$'s.
TWW
He wasn't; he hasn't.
All the Pixar films so far should have at least been nominated for Best Film of their year, I think the time when that can happen is moved further away by this category.
TWW