I think that the old games had a personality, an authentic voice, an author's vision, warmth, artistry, call it what you will. Part of good art is working within limits, and making those limits part of the work. We enjoy limericks not because they tell a great story or move us emotionally, but because of the author's skill in squeezing the rhyme into the five-line structure.
This is not to denigrate the equally skilled achievements of today's game developers in squeezing maximum performance from the 3D graphics hardware or ingeniously overcoming network latency problems. But a game written by a single developer is likely to have more individual personality than one by a team of hundreds, however impressive and photorealistic the graphics.
So your argument is that C compiles down to code which runs on the bare processor, whereas Perl etc. compiles to a bytecode that runs in a much richer runtime environment with garbage collection. This makes sense. I agree that C is not a high level language - just wanted to point out that 25 years ago, it was considered more of a high-level language than it is considered now.
I agree. Especially since path from source to execution can vary for the same language (Python can be compiled, and C interpreted, Haskell is both, and in any case there is usually some intermediate bytecode involved).
However I wouldn't use 'scripting' to just mean a language which is interpreted. For me, a scripting language is one where you can write small useful programs quickly - which suggests some kind of interpreter is available, it has dynamic typing, provides reasonable high-level data structures like lists and dictionaries, comes with useful libraries, etc.
C provides an abstract view of the machine in that you don't have to care about how many registers the processor has; you can just declare as many variables as you want and they will be allocated to registers automatically when needed for computation. It provides abstract data structures (structs) where you don't have to care about the memory layout. I agree that ML could be considered more abstract than C, and assembler less abstract; I don't think that 'provides an abstract view of the machine' is a good criterion, because it's too, um, abstract.
There's no particular requirement that C provide only integer or floating point types supported by the hardware. It is quite possible to have a full ANSI C implementation where the size of an int is not the processor's native integer.
There seem to be similarities between the Wikipedia cabal blocking all edits from a particular IP range and spam blacklist services that recommend blocking mail from a particular range. As Jef Poskanzer wrote:
Well, I don't know why, but in practice every single DNS-RBL eventually comes under the control of power-hungry weenies. They start listing sites unreliably, and if you complain you find yourself listed. And there's usually no way to get off the list.
Sound familiar? From TFA, it appears that Wikipedia blocked an IP range not because of abuse on Wikipedia, but because someone expressed his own views on his own private website. Similarly spam blacklists have been known to block people for 'promoting spam' by hosting web pages, even when those actions are not correlated with sending messages you'd want to block. Web filtering programs often block pages which are critical of web filtering, just for expressing an opinion the filtering company doesn't like, not for hosting obscene material.
Is there any way around the 'power-hungry weenie' problem? I think some explicit policy on blocking could help. If any IP address is blocked from Wikipedia, there must be a link to an archived copy of the Wikipedia vandalism that was responsible for the blocking, and this evidence should be verifiable by anyone.
Re:Perl 6: The Language of the Future (... Forever
on
State of the Onion 11
·
· Score: 1
If perl 6 had been released, Larry would be talking about perl 7.
Python, Perl, Lua etc. are very high level languages, at least from a slightly old-fashioned perspective that regards assembler as the baseline and something like C or Pascal as high level. But there are other very high level languages which are not scripting languages, for example Haskell, Erlang, Prolog, C++ with insane template nonsense, even SQL and XSLT. So the term scripting language is still useful I think. The definition is a bit fuzzy, which is what Larry points out in his talk.
The article's point is this. If the law is really so clear-cut, why the need to examine witnesses? The process followed by the judge suggests that the case hinges on facts and not just legal theory. If so, it should be tried by a jury.
I think when the other poster says 'expert' or 'expertise' he really means anyone with a basic knowledge of a subject and who isn't totally ignorant. Like a physics expert would be anyone with a university degree in physics. Articles on Wikipedia do need to be maintained by such 'experts' because someone who hasn't studied physics (or Chinese - or masonry - or ballet) wouldn't be able to add useful information or correct mistakes.
I think the answers to your questions about what constitutes damage are obvious. In any case, theft is already a crime under existing laws, and copyright infringement is already a civil wrong and/or a crime, so the most serious attacks could be dealt with even without any computer crime legislation at all.
Today it is practically impossible to do anything about attacks to home and small business computers. Local law enforcement doesn't have jurisdiction and the FBI wants hard proof of greater than $25,000 in losses.
Which seems to be an argument for reforming the law. I would rather have a relatively liberal set of laws, applied strictly, than a loose and overly broad law which is then applied selectively depending on what the FBI feel like doing.
At the moment if you find a vulnerability in a web site (even stumbling across it accidentally) there's no way you would ever report it. By doing so you open yourself up to prosecution and there is no reward for doing the right thing. If you find a vulnerability in a company's product you may be sued by that company under the DMCA or similar laws - better to keep it secret. I think that in these cases overly strict laws work against computer security.
With attempted 'hacking' from other countries, we see that domestic laws prohibiting unauthorized computer access are not much use. Of course they don't deter the Chinese army or any other government agency. They do deter domestic hackers, but have unpleasant side-effects like criminalizing viewing a page on a website to make sure it's not a phishing site. And if your computer security is oriented more towards tracking down individuals and bringing them to trial, you will be relatively defenceless against foreign agencies. Children brought up in an artificially clean and disinfected environment can suffer more infections when later exposed to the real world. It might be a better idea to legalize hacking, provided no damage is done, in order to strengthen your country's immune system.
I was more questioning the oddity of the original poster writing a message at all, because we don't normally post on Slashdot criticizing people's private lives.
In that case you can have no problem with every SUV owner on the planet.
As individuals, no. I do support measures to cap carbon emissions, which might include taxing gasoline more. I wouldn't single out one person who has an SUV.
Surely what they do with their own money is their own business. Presumably some people were happy to give money to the founders in exchange for shares of Google stock, and that's why the founders are rich now.
All the students need is a terminal account on a Linux box somewhere with mutt or pine (and IMAP). That will be more than adequate for the official email communication needed as part of their course. If for some unfathomable reason they'd rather use Outhouse or some other spam-ridden monstrosity instead of pine, leave them to it.
Microsoft's record of 'legal compliance' in any other area is remarkably poor; they tend to drag their feet on anything which would help interoperability and the consumer; but when it comes to misfeatures which restrict the user and reduce the PC's capability, they eagerly implement and gold-plate them. It would be good to see some of the robust Microsoft screw-you attitude applied to the movie studios and their ridiculous demands as well as to antitrust authorities.
The survey covers how many people are running a particular distribution and replied to the survey, not how many are running a particular distribution, and certainly not how many paid money for it. Novell and other companies are mostly interested in the third number, a bit in the second, and not in the first.
Customers are not people who downloaded a Linux distro gratis. They are people who have paid money for it.
It depends - accountants have odd requirements and often want to correct older things, except when they don't. After all, data entry errors do happen, and if you spot a mistake a week later, what are you going to do about it? Either you fix the original journal so you now have a correct picture of your finances, or you must resort to some bizarre workaround.
That's a good plan but it should be possible, whenever you want, to delete everything in the summary balance table and have it recalculated the next time it's wanted. Otherwise when you correct an error in an older journal the summary would then be incorrect.
I'm not saying that you really would delete all summaries, just that this is a good criterion for robustness of the system and shows it is always able to get back in sync.
Agreed, and never mind future needs, just for current ones having a normalized schema is the way to go. There isn't really any tradeoff between normalization and complexity. So far, every schema I've seen that was designed by some cluebie who decided to duplicate data 'to make it run faster' has ended up making code much more complex, because the same data is in several places and you have to remember to update all of them and what do you do if it gets out of sync (as it will)?
Granted, if you have at least five years' beard growth and experience with serious-sized databases (not a just few thousand records here and there) then you can denormalize it. Normally just for one thing, in one or two tables, and if there's a clear performance need you must meet.
Of course evolution on alien worlds will take a similar course to our own evolution. We can expect too that cultural development will follow broadly similar patterns; alien societies must surely discover fire, writing, the wheel, religion, and language following similar rules to our own. This explains why all aliens speak English. There must surely be an alien Slashdot where at this very moment alien nerds are having this very same discussion. And no doubt these alien people will have created science fiction series based on the assumption of parallel development just as we have, so their Star Trek will look just like our Star Trek, with possible minor changes, like Kirk having green skin and the Orion slave girl being human-coloured.
Indeed, with such similarity between aliens and ourselves, it may be pointless to ask whether alien species are already present here on Earth. We would not be able to tell the difference.
It depends what you understand by the spirit of the GPL. If you read the GPL version 2 that came out in 1991, the preamble makes it pretty clear:
the GNU General Public License is intended to guarantee your freedom to share and change free software--to make sure the software is free for all its users.
If all its users are to have the freedom to share and change the software, then that would include users who are accessing it over a web service. So this is a logical step to ensure freedom for software users.
Are you saying that if I, say, use an AGPL-licensed currency converting web service on my ticket reservation site, I must release the WHOLE site under the AGPL?
I don't think that is the case - just the source code to the currency converter itself, and any modifications you have made to that converter. Of course if you take the converter's source code and include it in your own code then you must be careful - but that is true with almost any software.
Slashdot Mirror
I think that the old games had a personality, an authentic voice, an author's vision, warmth, artistry, call it what you will. Part of good art is working within limits, and making those limits part of the work. We enjoy limericks not because they tell a great story or move us emotionally, but because of the author's skill in squeezing the rhyme into the five-line structure.
This is not to denigrate the equally skilled achievements of today's game developers in squeezing maximum performance from the 3D graphics hardware or ingeniously overcoming network latency problems. But a game written by a single developer is likely to have more individual personality than one by a team of hundreds, however impressive and photorealistic the graphics.
So your argument is that C compiles down to code which runs on the bare processor, whereas Perl etc. compiles to a bytecode that runs in a much richer runtime environment with garbage collection. This makes sense. I agree that C is not a high level language - just wanted to point out that 25 years ago, it was considered more of a high-level language than it is considered now.
I agree. Especially since path from source to execution can vary for the same language (Python can be compiled, and C interpreted, Haskell is both, and in any case there is usually some intermediate bytecode involved).
However I wouldn't use 'scripting' to just mean a language which is interpreted. For me, a scripting language is one where you can write small useful programs quickly - which suggests some kind of interpreter is available, it has dynamic typing, provides reasonable high-level data structures like lists and dictionaries, comes with useful libraries, etc.
I don't understand what you mean. Do you mean that C is a compiled language (generating machine code) while Python is interpreted?
C provides an abstract view of the machine in that you don't have to care about how many registers the processor has; you can just declare as many variables as you want and they will be allocated to registers automatically when needed for computation. It provides abstract data structures (structs) where you don't have to care about the memory layout. I agree that ML could be considered more abstract than C, and assembler less abstract; I don't think that 'provides an abstract view of the machine' is a good criterion, because it's too, um, abstract.
There's no particular requirement that C provide only integer or floating point types supported by the hardware. It is quite possible to have a full ANSI C implementation where the size of an int is not the processor's native integer.
Is there any way around the 'power-hungry weenie' problem? I think some explicit policy on blocking could help. If any IP address is blocked from Wikipedia, there must be a link to an archived copy of the Wikipedia vandalism that was responsible for the blocking, and this evidence should be verifiable by anyone.
If perl 6 had been released, Larry would be talking about perl 7.
Python, Perl, Lua etc. are very high level languages, at least from a slightly old-fashioned perspective that regards assembler as the baseline and something like C or Pascal as high level. But there are other very high level languages which are not scripting languages, for example Haskell, Erlang, Prolog, C++ with insane template nonsense, even SQL and XSLT. So the term scripting language is still useful I think. The definition is a bit fuzzy, which is what Larry points out in his talk.
The article's point is this. If the law is really so clear-cut, why the need to examine witnesses? The process followed by the judge suggests that the case hinges on facts and not just legal theory. If so, it should be tried by a jury.
I think when the other poster says 'expert' or 'expertise' he really means anyone with a basic knowledge of a subject and who isn't totally ignorant. Like a physics expert would be anyone with a university degree in physics. Articles on Wikipedia do need to be maintained by such 'experts' because someone who hasn't studied physics (or Chinese - or masonry - or ballet) wouldn't be able to add useful information or correct mistakes.
At the moment if you find a vulnerability in a web site (even stumbling across it accidentally) there's no way you would ever report it. By doing so you open yourself up to prosecution and there is no reward for doing the right thing. If you find a vulnerability in a company's product you may be sued by that company under the DMCA or similar laws - better to keep it secret. I think that in these cases overly strict laws work against computer security.
With attempted 'hacking' from other countries, we see that domestic laws prohibiting unauthorized computer access are not much use. Of course they don't deter the Chinese army or any other government agency. They do deter domestic hackers, but have unpleasant side-effects like criminalizing viewing a page on a website to make sure it's not a phishing site. And if your computer security is oriented more towards tracking down individuals and bringing them to trial, you will be relatively defenceless against foreign agencies. Children brought up in an artificially clean and disinfected environment can suffer more infections when later exposed to the real world. It might be a better idea to legalize hacking, provided no damage is done, in order to strengthen your country's immune system.
I was more questioning the oddity of the original poster writing a message at all, because we don't normally post on Slashdot criticizing people's private lives.
Surely what they do with their own money is their own business. Presumably some people were happy to give money to the founders in exchange for shares of Google stock, and that's why the founders are rich now.
Isn't the 767 bought by the founders of Google with their own money? It would be a bit fishy if bought by Google.
All the students need is a terminal account on a Linux box somewhere with mutt or pine (and IMAP). That will be more than adequate for the official email communication needed as part of their course. If for some unfathomable reason they'd rather use Outhouse or some other spam-ridden monstrosity instead of pine, leave them to it.
Microsoft's record of 'legal compliance' in any other area is remarkably poor; they tend to drag their feet on anything which would help interoperability and the consumer; but when it comes to misfeatures which restrict the user and reduce the PC's capability, they eagerly implement and gold-plate them. It would be good to see some of the robust Microsoft screw-you attitude applied to the movie studios and their ridiculous demands as well as to antitrust authorities.
The survey covers how many people are running a particular distribution and replied to the survey, not how many are running a particular distribution, and certainly not how many paid money for it. Novell and other companies are mostly interested in the third number, a bit in the second, and not in the first.
Customers are not people who downloaded a Linux distro gratis. They are people who have paid money for it.
It depends - accountants have odd requirements and often want to correct older things, except when they don't. After all, data entry errors do happen, and if you spot a mistake a week later, what are you going to do about it? Either you fix the original journal so you now have a correct picture of your finances, or you must resort to some bizarre workaround.
That's a good plan but it should be possible, whenever you want, to delete everything in the summary balance table and have it recalculated the next time it's wanted. Otherwise when you correct an error in an older journal the summary would then be incorrect.
I'm not saying that you really would delete all summaries, just that this is a good criterion for robustness of the system and shows it is always able to get back in sync.
Agreed, and never mind future needs, just for current ones having a normalized schema is the way to go. There isn't really any tradeoff between normalization and complexity. So far, every schema I've seen that was designed by some cluebie who decided to duplicate data 'to make it run faster' has ended up making code much more complex, because the same data is in several places and you have to remember to update all of them and what do you do if it gets out of sync (as it will)?
Granted, if you have at least five years' beard growth and experience with serious-sized databases (not a just few thousand records here and there) then you can denormalize it. Normally just for one thing, in one or two tables, and if there's a clear performance need you must meet.
Of course evolution on alien worlds will take a similar course to our own evolution. We can expect too that cultural development will follow broadly similar patterns; alien societies must surely discover fire, writing, the wheel, religion, and language following similar rules to our own. This explains why all aliens speak English. There must surely be an alien Slashdot where at this very moment alien nerds are having this very same discussion. And no doubt these alien people will have created science fiction series based on the assumption of parallel development just as we have, so their Star Trek will look just like our Star Trek, with possible minor changes, like Kirk having green skin and the Orion slave girl being human-coloured.
Indeed, with such similarity between aliens and ourselves, it may be pointless to ask whether alien species are already present here on Earth. We would not be able to tell the difference.
Also - that they were sending it by post at all instead of transferring it electronically (encrypted of course)...