1/ the BBC article in question uses a graphic which shows an NNTP client displaying the group alt.binaries.pictures.erotica.teen
Yes. The BBC, like slashdot, use 'category' related stock images for stories that they don't have a specific image for. This is their 'child porn' image.
There is an english seaside town names scunthorpe, because it contains the word CUNT in the name it is routinely blocked by world + dog using cheap filters, again we need to know what these filters consist of, if it is merely "teen" then it's bullshit, 19 years old porn queens abound...
It sounds like you glanced at TFA, but didn't read it. It's a list of sites that have been hand-vetted by some self appointed net-watchdog as being 'illegal'.
why can't they put in a message to warn of such illegal activities?
Because if they accidentally blocked an entirely innocent site, that could be construed as defamation or even libel, and land them with a rather expensive legal case.
The sites looked at are definitly illegal to look at.
Are you sure? How do you know that? Remember, only a judge & jury can decide what's illegal and what isn't.
The IWF's web site has a lot of language about "potentially illegal content", so they're obviously aware of the distinction. I wonder how many other people are.
Try following the link to that article, and come back again if you think that perl's inconsistencies are worse than PHPs.
I'm not the original poster, but that article is highly biased against PHP. Here are a few notes I've made on what is wrong with it:
Arguments and return values are extremely inconsistent
"In Perl, all the functionality provided by the functions in this table is available through a simple set of 4 operators"
The same is actually true of PHP: you can achieve the same results as any of these functions using only preg_match_all, preg_replace, strpos and stripos. However, the other functions may be easier to use and the code easier to read afterwards depending on what precisely you are trying to achieve. I'm not including the multibyte character string functions in this; until a variant of the preg_* functions is available that handles them, multibyte strings in PHP are broken, IMO.
PHP has separate functions for case insensitive operations
In Perl, you use a double lc() (lowercase) or the/i flag where PHP usually provides a case insensitive variant
You can do exactly the same in PHP, if you prefer doing it that way.
PHP has no lexical scope
True. However, in my 5 years' experience as a PHP software engineer, including a number of projects that I would class as 'large', this has actually turned out to be helpful, increasing my productivity. In a well designed system, where functionality is decomposed into numerous small modules, scoping issues tend to disappear.
PHP has too many functions in the core
They're comparing PHP with a large number of optional features included with the core PERL language. The comparison is clearly ludicrous; many of the features included in the PHP build used as reference require PERL modules or extensions to be present to have equivalent functions in PERL, these should have been present for the comparison.
Why does this always come up? I in particular hate the extremists on both sides of the fence. Why must everything be GPL compatible? It is a rhetorical question.
I'll answer it anyway. The problem is that PHP is a programming language interpreter that is designed to integrate with database software. MySQL is database software, licensed under the GPL, that PHP can integrate with and which is the preferred database software of a very large percentage of PHP users.
The GPL incorporates a clause that states that if you link GPL code to any other code and distribute the result, you must license the other code under the GPL (or, equivalently, some license that contains no restrictions that the GPL doesn't). This applies even to dynamic linking.
PHP doesn't do this, so consequently, nobody can legally distribute a compiled copy of PHP with the MySQL module compiled in.
This is a big problem that must be solved one way or another. One side of the fence believes that PHP should drop all of their license restrictions that aren't in the GPL, the other believes that MySQL's client library (the portion of MySQL that needs to be linked with PHP) should be distributed under a license that doesn't contain the restriction I described above (e.g. the LGPL).
Cheers for explaining the joke, none of us could get it, thank God for your intellect.
I wish somebody would explain it. What's so funny about a computer with a 133 year uptime? Or is this some new definition of "funny" that I have been previously unaware of?
The use of both printf and strlen in this case is absolutely fine. I've never heard of strlen leading to a security problem, and have rarely seen a non-trivial C program that doesn't use it. The only security problem I'm aware of with printf is that you shouldn't ever pass unchecked content to the format parameter, as this can lead to arbitrary code execution on some systems, apparently (I don't understand the precise mechanism involved).
So, either you're trolling against the entire C language, or are just ignorant.
I mean, it's not as if MS have never released software that stops working after a period of time (e.g. the PPP stack in Win98, which fails on a rather regular 28 day (or rather 2^31ms) cycle for me).
And I strongly suspect they do contract external testing companies, who might notice such an issue. And I also suspect that they have a penalty clause in their contract for any 'showstopper' bugs that should be missed.
So, what is it about this poster's claims you find so unlikely?
My gosh - the folks here who rabidly espouse the need for public outting of information all post anonymously.
No, we don't. There are plenty of us willing to identify ourselves.
I'll admit that I've never made an exploit like this public, although I have discovered a far more serious problem with a UK-based online credit card payment processor; unfortunately I am required by a contract of which I am signatory not to release any details of this problem, despite the fact that I suspect it was fixed years ago.
A quick google shows that it seems not to have become common knowledge, either.
But you can't sue them UNTIL your information is in the hand of someone who uses it illegally.
In the UK, the Data Protection Act 1984 requires anyone storing personal information on a computer system for anything other than personal use to take all reasonable steps to ensure that the data is secure. This means that you are wrong in this assertion -- you can initiate action against them if you can show that your information potentially could end up in the hands of someone who isn't authorised to receive it. The action isn't suing them, though, so you're partially right. The correct approach would be a complaint to the Information Commissioner, who would likely issue an order that the problems be fixed, followed by a fine if they are not after a reasonable period.
Morality, and by corollation, law and justice are not relative.
You state that as if it were an easily proved result, rather than the subject of many of the most heated debates of modern philosophy.
The general concensus is that morality is at least partially subjective. It is certainly true that there are many different moral systems throughout the world and the question of who can say which are 'right' and which are 'wrong' with authority is at least a difficult one to answer convincingly.
You also assert that the laws against trespass are a fundamental moral principle, while many cultures do not in fact have such a principle. In fact, the closest there is to a fundamental moral principle is "don't kill your friends (unless they want you to)", and I believe that even that isn't universally applied.
That said, the law is absolute (at least in most respects). This means that it is an attempt to write regulations that enforce "moral" behaviour (for some particular value of "moral" that is quite hard to decide). Of course, it is imperfect, as all such attempts must be -- at the very least the people deciding what is "moral" will change, and with them the definition of morality that is being used as the guide. In any modern society there is a very wide range of different moral beliefs. The law cannot encompass all of them.
Re:Not quite the same thing, but...
on
Biomorphic Software
·
· Score: 2, Informative
The best approach that I've found is to have a single thread for all timer related activities and set its affinity so that it always runs on the same processor. It simply waits on a semaphore and updates a global timestamp variable every time you signal it. It can also signal other semaphores after a specific delay (getting there with enough resolution might involve a bit of busy waiting, but typically for less than 2ms).
Yes, the main google index is only updated infrequently (every 2 weeks I think), so there aren't many NYT articles in there, only the ones that were current at the time an update happened and haven't been re-spidered yet.
The problem is simply that google _cannot_ index a site like the NYT. It takes anywhere from 3 to 30 days for a new page to get into google's index, and pages on the NYT site just don't last that long.
That's why google started their news section, so that they could do something with new pages that didn't survive long enough to make it into their main index.
Yes, the pages are in the archive, but google can't index the archive, because google would have to _pay_ to index the archive, and I don't think it's worth that much to them...
It's getting bad when even Reuters' journalists can't punctuate a sentence properly.
The program, which has never been tested fully, was launched after the Sept. 11, 2001, hijacking attacks to refine electronic techniques for using personal information to identify and rate potential threats.
This sentence means that the program hijacked attacks [...]. I think they meant:
The program, which has never been tested fully, was launched after the Sept. 11, 2001 hijacking attacks to refine electronic techniques for using personal information to identify and rate potential threats.
OK, here's my issues performance related with Mozilla (I'm on quite an old version now, 1.5, so maybe some of these have been fixed):
- startup time is slow, much worse than IE + Windows desktop load time (to account for the preloaded parts of IE).
- random pauses. Mozilla seems to occasionally stop responding for about.3 seconds. I suspect it might be using a garbage collector that has been badly tuned (?).
- html editing component (e.g. mail's compose window) has serious issues with long documents; IE's equivalent component is much faster, although not as nice IMHO.
Related problems:
- memory consumption is much higher than IE.
- some operations (e.g. moving a large volume of e-mail between mailboxes) seem to tie up all open Mozilla windows while they occur, which isn't very nice.
I'd submit these as RFEs on bugzilla, but my experience is that anything of bugzilla that isn't a showstopper bug just gets ignored for 2 years.
Well Dan Bricklin does point out that software of today can run on different hardware and having software tied to specific hardware is a bad idea
Software of today can run on a variety of different hardware, but there is a degree of similarity between the different types of hardware that probably won't exist between todays computers and those available a hundred years from today, much less two.
He is not just talking about one specific program that doesn't change, but rather open standards and techniques that mean data that is stored today, will be accessible in 200 years time.
That, on the other hand, I can agree with. Anyone storing information in a format that isn't publically documented really ought to consider whether they'll still need it in 30 years time, and start migrating it to an open format now if they will. However, there are very few formats that are completely undocumented. I believe the most commonly used might be Microsoft Access databases. I'm not sure what documentation exists on the formats of various other commercial database systems; I believe Oracle's formats are well documented (?). What about MSSQL? Informix?
Most accounts packages have documentation available on their database formats I believe. Certainly Sage and Pegasus provide such documentation. What about Great Plains, etc.?
Slashdot Mirror
1/ the BBC article in question uses a graphic which shows an NNTP client displaying the group alt.binaries.pictures.erotica.teen
Yes. The BBC, like slashdot, use 'category' related stock images for stories that they don't have a specific image for. This is their 'child porn' image.
There is an english seaside town names scunthorpe, because it contains the word CUNT in the name it is routinely blocked by world + dog using cheap filters, again we need to know what these filters consist of, if it is merely "teen" then it's bullshit, 19 years old porn queens abound...
It sounds like you glanced at TFA, but didn't read it. It's a list of sites that have been hand-vetted by some self appointed net-watchdog as being 'illegal'.
if it is usenet [...]
It isn't. RTFA.
why can't they put in a message to warn of such illegal activities?
Because if they accidentally blocked an entirely innocent site, that could be construed as defamation or even libel, and land them with a rather expensive legal case.
In the UK they are the same thing.
The sites looked at are definitly illegal to look at.
Are you sure? How do you know that? Remember, only a judge & jury can decide what's illegal and what isn't.
The IWF's web site has a lot of language about "potentially illegal content", so they're obviously aware of the distinction. I wonder how many other people are.
Try following the link to that article, and come back again if you think that perl's inconsistencies are worse than PHPs.
/i flag where PHP usually provides a case insensitive variant
I'm not the original poster, but that article is highly biased against PHP. Here are a few notes I've made on what is wrong with it:
Arguments and return values are extremely inconsistent
"In Perl, all the functionality provided by the functions in this table is available through a simple set of 4 operators"
The same is actually true of PHP: you can achieve the same results as any of these functions using only preg_match_all, preg_replace, strpos and stripos. However, the other functions may be easier to use and the code easier to read afterwards depending on what precisely you are trying to achieve. I'm not including the multibyte character string functions in this; until a variant of the preg_* functions is available that handles them, multibyte strings in PHP are broken, IMO.
PHP has separate functions for case insensitive operations
In Perl, you use a double lc() (lowercase) or the
You can do exactly the same in PHP, if you prefer doing it that way.
PHP has no lexical scope
True. However, in my 5 years' experience as a PHP software engineer, including a number of projects that I would class as 'large', this has actually turned out to be helpful, increasing my productivity. In a well designed system, where functionality is decomposed into numerous small modules, scoping issues tend to disappear.
PHP has too many functions in the core
They're comparing PHP with a large number of optional features included with the core PERL language. The comparison is clearly ludicrous; many of the features included in the PHP build used as reference require PERL modules or extensions to be present to have equivalent functions in PERL, these should have been present for the comparison.
Why does this always come up? I in particular hate the extremists on both sides of the fence. Why must everything be GPL compatible? It is a rhetorical question.
I'll answer it anyway. The problem is that PHP is a programming language interpreter that is designed to integrate with database software. MySQL is database software, licensed under the GPL, that PHP can integrate with and which is the preferred database software of a very large percentage of PHP users.
The GPL incorporates a clause that states that if you link GPL code to any other code and distribute the result, you must license the other code under the GPL (or, equivalently, some license that contains no restrictions that the GPL doesn't). This applies even to dynamic linking.
PHP doesn't do this, so consequently, nobody can legally distribute a compiled copy of PHP with the MySQL module compiled in.
This is a big problem that must be solved one way or another. One side of the fence believes that PHP should drop all of their license restrictions that aren't in the GPL, the other believes that MySQL's client library (the portion of MySQL that needs to be linked with PHP) should be distributed under a license that doesn't contain the restriction I described above (e.g. the LGPL).
So far, there is no movement.
Sorry, there's a long queue. I've been coding in PHP for years, and am still waiting.
Cheers for explaining the joke, none of us could get it, thank God for your intellect.
I wish somebody would explain it. What's so funny about a computer with a 133 year uptime? Or is this some new definition of "funny" that I have been previously unaware of?
I write this as I wait for fucking Acrobat to load
Downgrade to version 5. It doesn't have the bloat that's the problem with version 6.
One more thing:
The use of both printf and strlen in this case is absolutely fine. I've never heard of strlen leading to a security problem, and have rarely seen a non-trivial C program that doesn't use it. The only security problem I'm aware of with printf is that you shouldn't ever pass unchecked content to the format parameter, as this can lead to arbitrary code execution on some systems, apparently (I don't understand the precise mechanism involved).
So, either you're trolling against the entire C language, or are just ignorant.
I think that was the point - from the referenced page:
Splint is a tool for statically checking C programs for security vulnerabilities and coding mistakes
How do we know it?
I mean, it's not as if MS have never released software that stops working after a period of time (e.g. the PPP stack in Win98, which fails on a rather regular 28 day (or rather 2^31ms) cycle for me).
And I strongly suspect they do contract external testing companies, who might notice such an issue. And I also suspect that they have a penalty clause in their contract for any 'showstopper' bugs that should be missed.
So, what is it about this poster's claims you find so unlikely?
My gosh - the folks here who rabidly espouse the need for public outting of information all post anonymously.
No, we don't. There are plenty of us willing to identify ourselves.
I'll admit that I've never made an exploit like this public, although I have discovered a far more serious problem with a UK-based online credit card payment processor; unfortunately I am required by a contract of which I am signatory not to release any details of this problem, despite the fact that I suspect it was fixed years ago.
A quick google shows that it seems not to have become common knowledge, either.
But you can't sue them UNTIL your information is in the hand of someone who uses it illegally.
In the UK, the Data Protection Act 1984 requires anyone storing personal information on a computer system for anything other than personal use to take all reasonable steps to ensure that the data is secure. This means that you are wrong in this assertion -- you can initiate action against them if you can show that your information potentially could end up in the hands of someone who isn't authorised to receive it. The action isn't suing them, though, so you're partially right. The correct approach would be a complaint to the Information Commissioner, who would likely issue an order that the problems be fixed, followed by a fine if they are not after a reasonable period.
I got thrown out of the school library for taking the mice apart to clean the balls.
Of course, once people saw me doing that, everyone started taking the balls out and throwing them at each other...
Morality, and by corollation, law and justice are not relative.
You state that as if it were an easily proved result, rather than the subject of many of the most heated debates of modern philosophy.
The general concensus is that morality is at least partially subjective. It is certainly true that there are many different moral systems throughout the world and the question of who can say which are 'right' and which are 'wrong' with authority is at least a difficult one to answer convincingly.
You also assert that the laws against trespass are a fundamental moral principle, while many cultures do not in fact have such a principle. In fact, the closest there is to a fundamental moral principle is "don't kill your friends (unless they want you to)", and I believe that even that isn't universally applied.
That said, the law is absolute (at least in most respects). This means that it is an attempt to write regulations that enforce "moral" behaviour (for some particular value of "moral" that is quite hard to decide). Of course, it is imperfect, as all such attempts must be -- at the very least the people deciding what is "moral" will change, and with them the definition of morality that is being used as the guide. In any modern society there is a very wide range of different moral beliefs. The law cannot encompass all of them.
The best approach that I've found is to have a single thread for all timer related activities and set its affinity so that it always runs on the same processor. It simply waits on a semaphore and updates a global timestamp variable every time you signal it. It can also signal other semaphores after a specific delay (getting there with enough resolution might involve a bit of busy waiting, but typically for less than 2ms).
Yes, the main google index is only updated infrequently (every 2 weeks I think), so there aren't many NYT articles in there, only the ones that were current at the time an update happened and haven't been re-spidered yet.
The problem is simply that google _cannot_ index a site like the NYT. It takes anywhere from 3 to 30 days for a new page to get into google's index, and pages on the NYT site just don't last that long.
That's why google started their news section, so that they could do something with new pages that didn't survive long enough to make it into their main index.
Yes, the pages are in the archive, but google can't index the archive, because google would have to _pay_ to index the archive, and I don't think it's worth that much to them...
It's getting bad when even Reuters' journalists can't punctuate a sentence properly.
The program, which has never been tested fully, was launched after the Sept. 11, 2001, hijacking attacks to refine electronic techniques for using personal information to identify and rate potential threats.
This sentence means that the program hijacked attacks [...]. I think they meant:
The program, which has never been tested fully, was launched after the Sept. 11, 2001 hijacking attacks to refine electronic techniques for using personal information to identify and rate potential threats.
OK, here's my issues performance related with Mozilla (I'm on quite an old version now, 1.5, so maybe some of these have been fixed):
.3 seconds. I suspect it might be using a garbage collector that has been badly tuned (?).
- startup time is slow, much worse than IE + Windows desktop load time (to account for the preloaded parts of IE).
- random pauses. Mozilla seems to occasionally stop responding for about
- html editing component (e.g. mail's compose window) has serious issues with long documents; IE's equivalent component is much faster, although not as nice IMHO.
Related problems:
- memory consumption is much higher than IE.
- some operations (e.g. moving a large volume of e-mail between mailboxes) seem to tie up all open Mozilla windows while they occur, which isn't very nice.
I'd submit these as RFEs on bugzilla, but my experience is that anything of bugzilla that isn't a showstopper bug just gets ignored for 2 years.
Interesting article. I don't think he's necessarily right in all aspects of that, but he has some good ideas.
It's clear that he was approaching the question with the LISP-user's mindset: simpler is better. There is such a thing as too simple, in my opinion.
Well Dan Bricklin does point out that software of today can run on different hardware and having software tied to specific hardware is a bad idea
Software of today can run on a variety of different hardware, but there is a degree of similarity between the different types of hardware that probably won't exist between todays computers and those available a hundred years from today, much less two.
He is not just talking about one specific program that doesn't change, but rather open standards and techniques that mean data that is stored today, will be accessible in 200 years time.
That, on the other hand, I can agree with. Anyone storing information in a format that isn't publically documented really ought to consider whether they'll still need it in 30 years time, and start migrating it to an open format now if they will. However, there are very few formats that are completely undocumented. I believe the most commonly used might be Microsoft Access databases. I'm not sure what documentation exists on the formats of various other commercial database systems; I believe Oracle's formats are well documented (?). What about MSSQL? Informix?
Most accounts packages have documentation available on their database formats I believe. Certainly Sage and Pegasus provide such documentation. What about Great Plains, etc.?
The UK equivalent would be the Disability Discrimination Act (2002, I think), and contains similar provisions.
Here I have posted the equivalent UK law, which is somewhat easier to comply with.