Haven't I seen all this before?
on
Perl 6 Showcase
·
· Score: 3
20 years ago, there was a great programming language called "Lisp". It had lexical scoping, was both interpreted and compiled, allowed its syntax to be extended arbitrarily, had optional type declarations, and had multiple syntactic front-ends written in itself. In fact, its credo was to let people write "domain specific 'little' languages".
So, this leaves me wondering: why should Perl succeed where Lisp failed? What distinguishes the Perl6 goals from those of Lisp? And why start from scratch? Why not use use one of the many CommonLisp implementations, implementations that have already addressed most of the hard problems (generational garbage collection, native code generation, compilation to C, debuggers, etc.)?
The journalling file systems that are out there for Linux and UNIX log structural information only, and only temporarily. They allow recovery of directory structure in case of a small class of software/hardware failures (mostly, shutting down without flushing buffers), not much more. File content isn't logged, it isn't recoverable, and you can't roll back your file system arbitrarily.
Sure, you can implement something like tripwire (or various forms of capabilities) at the file system level. But that doesn't need to be anywhere near as complex as a JFS. You're probably better off with a simple, general change notification mechanism and a user-level daemon.
This may not beat the PS2 or DC for the "hardcore" console gamers, but it may be a good way of getting Linux into the home. If the price is reasonable, I'd certainly buy one as an MP3 player and for other purposes.
Well, the IA-64 is not only a messy instruction set to generate code for, and not only does its single existing implementation run like molasses, now other companies will also be prevented from coming up with compatible (and hopefully faster) implementations. I think the future of IA-64 is pretty uncertain...
Objective-C wasn't developed by the folks who did the Mach kernel. Furthermore, an Objective-C method call is roughly the same speed as a C function call in a good implementation.
Objective-C has a few problems inherited from pre-ANSI C, but those would be fixable. I think it would be a good language to write kernel components in. IMO, it would certainly beat the mess of dispatch tables and dynamic loading hacks currently found in the various operating system kernels based on C.
Instead of the Hurd, why not start with Darwin? Both the Hurd and Darwin derive from the same open CMU project. Darwin probably has seen a bit more use and bug fixing than the Hurd.
The design of the Mach kernel also should make it easier for companies to put in their favorite pet "enterprise" features (LVM, JFS, whatever) without messing up the system for everybody else.
Re:Some of us worked our way through art school...
on
Life as Video Game Art
·
· Score: 1
I pay for you or your kids to go to school, the roads you drive your car on, etc., and you pay for the art I enjoy. And it all goes through the government.
Sounds terribly inefficient? Not everybody is getting a fair deal? True, all true.
But eliminating the government from the equation and transferring it all to the free market is worse. If you don't believe me, there are lots of countries that work that way that you can visit at your leisure. I predict you'll be back soon.
While it may be inconvenient to have one's lecture notes published by note taking companies, and while some students may be encouraged to slack off by the availability of lecture notes on line, the potential of this legislation go much further.
Teaching itself is about passing on information across generations. Many students go on to become teachers themselves, and part of that is on passing on the teaching traditions of their own teachers. That, too, constitutes a use of "lecture notes" for commercial purposes, and if copyright is extended in this way, it threatens the continuity of teaching.
To me, this looks like a very short-sighted piece of legislation. Text book companies and professors may like it because it allows them to publish lecture notes as overpriced text books. But the long term implications, however, are very troubling to me since it affects the basis of our culture: the continuity of teaching across generations.
Both your views of student visas and your views of H1B visas are unsupported by facts. Student visas have no restrictions on continued stay; if their intent was as you say, they would be subject to the same home residency requirement as J1 visas. Furthermore, given that many foreign students are liberally supported by US institutions, it would be silly to send them away.
H1B visas were explicitly changed six years ago to become a first step for skill-based immigration. And they are generally not "slave labor"--H1B workers are generally paid at equivalent levels to US workers.
The primary problem with the current immigration system is the inefficient and inequitable administration by the INS. And US citizens should be up in arms about that because the US benefits greatly from the foreign workers that come here.
On Windows, there are so many "features" that hang off the native widgets, events, and windows, that I would be concerned about using an emulation. For just getting a nice looking UI, it may be good enough. But if you end up having to do something windows-specific (conditionalized for the Windows platform), you may not be able to do it.
I have had fairly good experience with wxWindows. It is quite close to MFC, so Windows programmers should feel at home, and even for non-Windows programmers, it's a good choice for GTK programming in C++. wxWindows has pretty much all the widgets and functionality you would expect in a current C++ toolkit. wxWindows also has an almost identical Python interface, wxPython, which means that you can easily switch back and forth between C++ and Python, as well as between Windows and Linux.
wxWindows still has some rough spots, but then so do almost all other toolkits. If you don't need quite as much functionality and can do without native LAF, FLTK is also pretty good and much smaller.
It's not clear that a video demonstration at some conference necessarily constitutes prior art. In order for something to be prior art, it needs to meet fairly stringent publication requirements.
The rationale is that the patent system exists to support the dissemination of ideas. If people are protected from patent claims merely because they did something first, the incentive for people to publish their ideas would be greatly diminished. With patents, so the legal thinking goes, it may take 21 years for it to become public domain but at least it will.
The lesson is: if you do something, even for an open source project, document it in writing and disclose it in proper form. You can disclose directly to the patent office without filing a patent.
None of the desktop environments, Gnome, KDE, XFce, and others, address what I really want in a desktop environment: an easy way to integrate applications and reuse functionality. Instead, they are all complex C/C++ programs with even more complex APIs.
Oh, sure, when I have to, I can hack them (I have written some Gnome/GTK code). But it feels like it's a lot more work than it ought to be. Let's hope some future desktop will address that issue.
Well, calling it "top secret" makes it all sound terribly important. But the reality is that their "top secret filtering technology" is probably no better than what a smart graduate student in information retrieval could produce in a couple of weeks. And their encryption is unlikely to be any better than what is publically known.
The NSA fell for the "security through obscurity" trap. I think the only thing that classifying all that stuff does is that it makes them feel very important, an exclusive club.
What they are doing isn't bad because of some nebulous belief "information is sacred", it's bad because it's stupid.
Even if the names had been blocked out correctly, they are publishing resumes with incredible detail. What is the point of blocking out the names?
How many "Division Managers, IIT Research Institute, Technology Assessment Division" can there be? How many "Professors of Law, Villanova University, 1981-1997"? Any skilled journalist or researcher could find the actual names of these people in a day, and if it matters would publish them.
The obvious conclusion is that the people involved in Carnivore are both technically incompetent (didn't they learn from the last PDF leak?) and apparently don't give a damn about even the simplest concepts about privacy.
But why get upset? It's the incompetent trying to catch the stupid. After all, Carnivore simply doesn't make sense in the presence of cryptography, remailers, web based message systems, and other widely available technologies. Anybody with even a miniscule amount of sense would know better than to send any information the FBI might be interested in through open E-mail.
There is no security. While it may be less convenient, you can access the same data from a boot floppy with "strings/dev/hda3", usually good enough to find passwords and personal information.
The purpose of trademarks is neither to protect the public image of companies, nor to protect their "goodwill" or "reputation" (we have product disparagement and libel laws for that). Trademarks aren't there to let companies make money by staking out claims to pieces of our language.
Trademarks exist for the benefit of consumers, to let them identify specific products unambiguously,
both to associate positive and to associate negative attributes with the product. If a trademark lets a company legally protect its reputation and "goodwill", then it is only by delivering a good product and behaving responsibly.
The use on F*ckedCompany seems like an entirely appropriate use of the trademark, in line with the intent of trademark law: it unambiguously identifies which company a particular notice refers to.
Trademark law explicitly concerns only the likelihood "to cause confusion, to cause mistake, or to deceive". If you buy a "Coca Cola", you are not supposed to be getting some other product. But that doesn't protect Coca Cola Company from negative statements (as long as they are truthful). In fact, trademarks are just as important for identifying bad products as they are for identifying good products.
That's the wrong approach. The Internet infrastructure itself should never let forged SYN packets out to begin with.
If we design our protocols to deal with such screwy cases, we are sacrificing a lot of performance. The initial roundtrip for TCP is one of the things that makes TCP such a lousy protocol for the web, as well as for many other "reliable datagram" applications. If the initial packet not only contained the identity of the source but also data (e.g., the request), TCP could be nearly as efficient as UDP for many applications.
If you want some kind of method that establishes that a remote host exists, at least don't do it on every packet. Instead, exchange some cryptographic tokens with only the first connection and then allow the remote host to use those tokens for communication without a roundtrip for future communications.
By using language like "the I.O.C.'s intellectual property rights", people are already conceding everything to the I.O.C. How could anybody oppose giving other people their property rights?
Of course, the term "intellectual property rights" is a cynical manipulation of language. Why should anybody have "property rights" to other people's free speech?
As for the Olympic "movement" itself, I hope athletes and viewer will stop deluding themselves about it. The Olympics is all about money. International understanding and sports are not a part of it. If you want to further the so-called "Olympic ideals", participate in sports yourself and travel yourself. There are lots of fun sports events around the world that are not as commercialized.
(Similar comments apply to the term "intellectual property rights" for trademarks, patents, and copyrights. All of those are limited licenses granted by the government, not intrinsic unlimited rights to property.)
In the current climate, poor security, insurance, and off-loading finding the criminals on overworked police is going to be cheaper for many E-commerce sites than hiring the right people to implement good security procedures.
Economics comes to the rescue: we should force companies to take full responsibility for the losses and the suffering that their carelessness causes others. Any company that exposes private data should be penalized, whether or not any losses are actually incurred by an individual. And if that company becomes target of a computer crime, they should not be able to hide behind someone else's guilt: they should be subject to stiff penalties. That way, poor security is not going to be the most rational economic choice for a company.
Going after the "computer criminals" is not going to be a solution. The current approach is only effective at catching bragging teenagers and allowing companies to shut up white-hat hackers; it won't catch anybody who is really determined to do damage and hide their identity: there are too many ways of using the Internet anonymously, from trial dial-up accounts and public phones to public Internet terminals. It has no force as a deterrent.
The only way consumer confidence can be created is by holding the people who can do something meaningful about security, the Internet companies who hold private data, responsible.
While UV is bad in lots of ways, there is a much simpler way in which looking at the sun hurts your eyes: heating. The sunlight gets focussed into a tiny spot, it heats up and cooks your retina in one spot.
You probably won't feel any pain, and the resulting damage is not directly visible--it's just like another blind spot. But once a significant part of your fovea (central vision) is damaged, you won't be able to see details anymore.
So, in short, don't look at the sun directly under any circumstances, and get regular eye exams. There are lots of other things that can go wrong with your eyes.
An emulator is not usually piracy,
it is an independently created piece of software. It is
debatable whether downloading ROMs is piracy; copyright law
has significant fair use exemptions for non-commercial copying,
in particular if it occurs among friend and families.
As for feeling sorry for game companies, there are several
things wrong with his arguments.
First, the more protection you give game
companies, the more likely they are going to reissue and rehash
and old game. Short term protection is necessary to give those
companies the ability to make a profit. But they have already
made huge profits. Long term protection, however, removes the incentive for game companies to innovate and come up with new game ideas.
In addition, copyright law was not intended to let companies
manipulate the market by letting titles go out of print and
then reissue them in a new format that requires technically
incompatible hardware and forces everybody to pay for the
same content many times over. Such behavior constitutes an
abuse of copyright law.
As for defending the companies that brought us all those good game console games--what about their effect on open source gaming? Open source games existed before the PC and consoles, but they have suffered significantly from the PC and console game companies' inroads. So, let's keep the "sympathy" angle out of this argument: it is hardly a clear win for Sega and the other game companies.
What this comes down to is that Sega and other game companies have already taken advantage of copyright laws. And they want even more: prohibitions on emulators, prohibitions on fair use, prohibitions on changes in storage medium, etc. But what's their justification? I don't see much interesting innovation. We, the people, decide what copyright law ought to be, to encourage creative talent to work. If we go the direction that Sega and its defenders want us to go in, we will stifle innovation and creativity for the financial gain of a few stockholders. I think we need to go the other direction: shorten copyrights and liberalized copyright laws. That would encourage innovation and result in better games in the long run, while still allowing authors to make a profit.
If Microsoft only were that blatant. But reality is that Outlook 2000 messages, for the most part, can be read in other mail readers, whether they have a TNEF attachment or not. Also, Microsoft changed ASCII years ago, but the changes were subtle: that's why you get the occasional "?" instead of various quotes and trademark symbols when reading MS-authored web pages.
If Microsoft software just didn't work with other mail readers, it would disrupt the network effects that they rely on for market dominance. They want other people to receive Outlook messages, they just don't want those messages to look quite right (plain text instead of RTF) or function quite right (no scheduling, etc.). That way, they will spread their message to the maximum number of people: "yes, you can read your mail, but it would look and work so much better if you, too, switched to Windows and Outlook".
The same is true for their web authoring. They can't afford to cut out 30% of the market that still use something other than IE, they just don't want the experience to be really good for those other people.
Areas where Microsoft can and has created much more complete incompatibility is with Word and Excel documents. In the past, those weren't sent around that much: they usually stayed within a single company. Now, as they are being used more for other purposes and distributed by mail and over the Internet, they are opening the format up just enough (by going to XML): "Yes, you can kind of read them with other tools, but you'll get less functionality. Isn't it time to switch to Word?"
There are often significant fixed expenses and risks associated with getting (and later losing) a new cutomer. This is particularly true for cell phone companies, where they company actually partially pays for the cell phone. But it's true (at least in an accounting sense) for ISPs and other businesss. Fixed term contracts are usually designed to account for those expenses in some reasonable way so that customer don't get hit with a large up-front expense.
Of course, as a customer, you should be thinking of that 1 year cell phone contract not as $30/month plus a $100 cell phone, but as a $460 purchase with a convenient payment plan. And once the costs for the phone have been recouped by the company, they are turning a significant profit.
A big problem with such contracts is that they remove much of the incentive for a company to provide good service. If service quality deteriorates during that time, customers still can't easily leave. So, bad service won't affect in the bottom line right away, and the usual feedback look between profits and quality is broken. In a sense, companies that write these kinds of contracts may be doing themselves a disservice, because they get much less information about their product from their customers.
Slashdot Mirror
So, this leaves me wondering: why should Perl succeed where Lisp failed? What distinguishes the Perl6 goals from those of Lisp? And why start from scratch? Why not use use one of the many CommonLisp implementations, implementations that have already addressed most of the hard problems (generational garbage collection, native code generation, compilation to C, debuggers, etc.)?
Maybe someone can explain to me what I'm missing.
Sure, you can implement something like tripwire (or various forms of capabilities) at the file system level. But that doesn't need to be anywhere near as complex as a JFS. You're probably better off with a simple, general change notification mechanism and a user-level daemon.
This may not beat the PS2 or DC for the "hardcore" console gamers, but it may be a good way of getting Linux into the home. If the price is reasonable, I'd certainly buy one as an MP3 player and for other purposes.
Well, the IA-64 is not only a messy instruction set to generate code for, and not only does its single existing implementation run like molasses, now other companies will also be prevented from coming up with compatible (and hopefully faster) implementations. I think the future of IA-64 is pretty uncertain...
Objective-C has a few problems inherited from pre-ANSI C, but those would be fixable. I think it would be a good language to write kernel components in. IMO, it would certainly beat the mess of dispatch tables and dynamic loading hacks currently found in the various operating system kernels based on C.
The design of the Mach kernel also should make it easier for companies to put in their favorite pet "enterprise" features (LVM, JFS, whatever) without messing up the system for everybody else.
Sounds terribly inefficient? Not everybody is getting a fair deal? True, all true.
But eliminating the government from the equation and transferring it all to the free market is worse. If you don't believe me, there are lots of countries that work that way that you can visit at your leisure. I predict you'll be back soon.
Teaching itself is about passing on information across generations. Many students go on to become teachers themselves, and part of that is on passing on the teaching traditions of their own teachers. That, too, constitutes a use of "lecture notes" for commercial purposes, and if copyright is extended in this way, it threatens the continuity of teaching.
To me, this looks like a very short-sighted piece of legislation. Text book companies and professors may like it because it allows them to publish lecture notes as overpriced text books. But the long term implications, however, are very troubling to me since it affects the basis of our culture: the continuity of teaching across generations.
H1B visas were explicitly changed six years ago to become a first step for skill-based immigration. And they are generally not "slave labor"--H1B workers are generally paid at equivalent levels to US workers.
The primary problem with the current immigration system is the inefficient and inequitable administration by the INS. And US citizens should be up in arms about that because the US benefits greatly from the foreign workers that come here.
On Windows, there are so many "features" that hang off the native widgets, events, and windows, that I would be concerned about using an emulation. For just getting a nice looking UI, it may be good enough. But if you end up having to do something windows-specific (conditionalized for the Windows platform), you may not be able to do it.
wxWindows still has some rough spots, but then so do almost all other toolkits. If you don't need quite as much functionality and can do without native LAF, FLTK is also pretty good and much smaller.
The rationale is that the patent system exists to support the dissemination of ideas. If people are protected from patent claims merely because they did something first, the incentive for people to publish their ideas would be greatly diminished. With patents, so the legal thinking goes, it may take 21 years for it to become public domain but at least it will.
The lesson is: if you do something, even for an open source project, document it in writing and disclose it in proper form. You can disclose directly to the patent office without filing a patent.
Oh, sure, when I have to, I can hack them (I have written some Gnome/GTK code). But it feels like it's a lot more work than it ought to be. Let's hope some future desktop will address that issue.
The NSA fell for the "security through obscurity" trap. I think the only thing that classifying all that stuff does is that it makes them feel very important, an exclusive club.
What they are doing isn't bad because of some nebulous belief "information is sacred", it's bad because it's stupid.
How many "Division Managers, IIT Research Institute, Technology Assessment Division" can there be? How many "Professors of Law, Villanova University, 1981-1997"? Any skilled journalist or researcher could find the actual names of these people in a day, and if it matters would publish them.
The obvious conclusion is that the people involved in Carnivore are both technically incompetent (didn't they learn from the last PDF leak?) and apparently don't give a damn about even the simplest concepts about privacy.
But why get upset? It's the incompetent trying to catch the stupid. After all, Carnivore simply doesn't make sense in the presence of cryptography, remailers, web based message systems, and other widely available technologies. Anybody with even a miniscule amount of sense would know better than to send any information the FBI might be interested in through open E-mail.
There is no security. While it may be less convenient, you can access the same data from a boot floppy with "strings /dev/hda3", usually good enough to find passwords and personal information.
Trademarks exist for the benefit of consumers, to let them identify specific products unambiguously, both to associate positive and to associate negative attributes with the product. If a trademark lets a company legally protect its reputation and "goodwill", then it is only by delivering a good product and behaving responsibly.
The use on F*ckedCompany seems like an entirely appropriate use of the trademark, in line with the intent of trademark law: it unambiguously identifies which company a particular notice refers to.
Trademark law explicitly concerns only the likelihood "to cause confusion, to cause mistake, or to deceive". If you buy a "Coca Cola", you are not supposed to be getting some other product. But that doesn't protect Coca Cola Company from negative statements (as long as they are truthful). In fact, trademarks are just as important for identifying bad products as they are for identifying good products.
If we design our protocols to deal with such screwy cases, we are sacrificing a lot of performance. The initial roundtrip for TCP is one of the things that makes TCP such a lousy protocol for the web, as well as for many other "reliable datagram" applications. If the initial packet not only contained the identity of the source but also data (e.g., the request), TCP could be nearly as efficient as UDP for many applications.
If you want some kind of method that establishes that a remote host exists, at least don't do it on every packet. Instead, exchange some cryptographic tokens with only the first connection and then allow the remote host to use those tokens for communication without a roundtrip for future communications.
Of course, the term "intellectual property rights" is a cynical manipulation of language. Why should anybody have "property rights" to other people's free speech?
As for the Olympic "movement" itself, I hope athletes and viewer will stop deluding themselves about it. The Olympics is all about money. International understanding and sports are not a part of it. If you want to further the so-called "Olympic ideals", participate in sports yourself and travel yourself. There are lots of fun sports events around the world that are not as commercialized.
(Similar comments apply to the term "intellectual property rights" for trademarks, patents, and copyrights. All of those are limited licenses granted by the government, not intrinsic unlimited rights to property.)
Economics comes to the rescue: we should force companies to take full responsibility for the losses and the suffering that their carelessness causes others. Any company that exposes private data should be penalized, whether or not any losses are actually incurred by an individual. And if that company becomes target of a computer crime, they should not be able to hide behind someone else's guilt: they should be subject to stiff penalties. That way, poor security is not going to be the most rational economic choice for a company.
Going after the "computer criminals" is not going to be a solution. The current approach is only effective at catching bragging teenagers and allowing companies to shut up white-hat hackers; it won't catch anybody who is really determined to do damage and hide their identity: there are too many ways of using the Internet anonymously, from trial dial-up accounts and public phones to public Internet terminals. It has no force as a deterrent.
The only way consumer confidence can be created is by holding the people who can do something meaningful about security, the Internet companies who hold private data, responsible.
You probably won't feel any pain, and the resulting damage is not directly visible--it's just like another blind spot. But once a significant part of your fovea (central vision) is damaged, you won't be able to see details anymore.
So, in short, don't look at the sun directly under any circumstances, and get regular eye exams. There are lots of other things that can go wrong with your eyes.
As for feeling sorry for game companies, there are several things wrong with his arguments.
First, the more protection you give game companies, the more likely they are going to reissue and rehash and old game. Short term protection is necessary to give those companies the ability to make a profit. But they have already made huge profits. Long term protection, however, removes the incentive for game companies to innovate and come up with new game ideas.
In addition, copyright law was not intended to let companies manipulate the market by letting titles go out of print and then reissue them in a new format that requires technically incompatible hardware and forces everybody to pay for the same content many times over. Such behavior constitutes an abuse of copyright law.
As for defending the companies that brought us all those good game console games--what about their effect on open source gaming? Open source games existed before the PC and consoles, but they have suffered significantly from the PC and console game companies' inroads. So, let's keep the "sympathy" angle out of this argument: it is hardly a clear win for Sega and the other game companies.
What this comes down to is that Sega and other game companies have already taken advantage of copyright laws. And they want even more: prohibitions on emulators, prohibitions on fair use, prohibitions on changes in storage medium, etc. But what's their justification? I don't see much interesting innovation. We, the people, decide what copyright law ought to be, to encourage creative talent to work. If we go the direction that Sega and its defenders want us to go in, we will stifle innovation and creativity for the financial gain of a few stockholders. I think we need to go the other direction: shorten copyrights and liberalized copyright laws. That would encourage innovation and result in better games in the long run, while still allowing authors to make a profit.
If Microsoft software just didn't work with other mail readers, it would disrupt the network effects that they rely on for market dominance. They want other people to receive Outlook messages, they just don't want those messages to look quite right (plain text instead of RTF) or function quite right (no scheduling, etc.). That way, they will spread their message to the maximum number of people: "yes, you can read your mail, but it would look and work so much better if you, too, switched to Windows and Outlook".
The same is true for their web authoring. They can't afford to cut out 30% of the market that still use something other than IE, they just don't want the experience to be really good for those other people.
Areas where Microsoft can and has created much more complete incompatibility is with Word and Excel documents. In the past, those weren't sent around that much: they usually stayed within a single company. Now, as they are being used more for other purposes and distributed by mail and over the Internet, they are opening the format up just enough (by going to XML): "Yes, you can kind of read them with other tools, but you'll get less functionality. Isn't it time to switch to Word?"
Of course, as a customer, you should be thinking of that 1 year cell phone contract not as $30/month plus a $100 cell phone, but as a $460 purchase with a convenient payment plan. And once the costs for the phone have been recouped by the company, they are turning a significant profit.
A big problem with such contracts is that they remove much of the incentive for a company to provide good service. If service quality deteriorates during that time, customers still can't easily leave. So, bad service won't affect in the bottom line right away, and the usual feedback look between profits and quality is broken. In a sense, companies that write these kinds of contracts may be doing themselves a disservice, because they get much less information about their product from their customers.