I wrote: Same end result, two different mechanisms. One of which you favor.
Then Eric wrote: Building added functionality into the operating system is not the issue. The issue is Microsoft coming out with their own version of a proxy server and then threatening Dell that they are going to withold their Windows license if they don't preload their server and only their server along with Windows. Or if they call up Apple and say "we'll cancel Mac Office if you don't preload our Mac proxy server, and deep six your own version for Windows."
But what does this have to do with the post I responded to? Answer: absolutely nothing.
Take internet connection sharing for example. I've gone through wingate, sygate, and finally winroute, all shareware applications operating in a viable market. Now microsoft in windows 98 second edition has come out with 'internet connection sharing' - which makes the entire market defunct!!! Microsoft is going around squashing honest programmers and companies.
So what would you do if I'd ported the same functionality from Linux over to Windows and released it under the GPL? Probably nothing. You wouldn't be jumping up and down about it.
Same end result, two different mechanisms. One of which you favor.
"For while consumers might one day turn to network computers, or Linux, or a combination of middleware and some other operating system, as an alternative to Windows, the fact remains that they are not doing so today. Nor are consumers likely to do so in appreciable numbers any time in the next few years." - Judge Jackson.
Welcome to the ISO-standardized mechanism for file handling.
1. Create temporary filename. 2. Create temporary file. 3. Copy entire file into another temporary file. 4. Dance on that file for a while. 5. Delete Original File. 6. Rename Temp file to Original filename. 7. Touch the file to update the file modification date and creation dates to match the original file.
Failure to use this mechanism will result in a slow and painful death, as administered by Nick, because as we all know, in computing there's only one TRUE way of doing anything, and anything other than the way you know and understand is obviously (1) inferior, (2) error prone, and (3) the spawn of Satan.
Pending Standards: Mouse button click duration; accurate to the nearest millisecond. Button-clicks by users outside of this 1ms time window will be ignored.
Standard function names. All function names must begin with the word "Fondue" for no apparent reason, other than because when I started programming 9 years ago, it seemed like a good idea at the time.
All programs must be written in C. No other languages are allowed.
When performing large file updates, you must create a delta file, which you then run through and create a "child" file from, by comparing the delta file to the parent file. You then must backup the parent file. You must perform all actions as three separate processes: add new records, sort, remove duplicates. Preferably overnight, when most people won't be using the mainframe time. (Oh sorry - home users don't have mainframes? Well, tough titty).
I became aware of another potentially anti-competitive thing MS did some years back, which, as far as I am aware no one else has mentioned in the press before. If it wasn't anti-competitive, it was at least stupid, and MS users are the ultimate victims.
You should get out more. You'll be telling me that it's anticompetitive to support HTML 4.0 in web documents soon, because "Mosaic 1.0 doesn't support it".
umm.. i dont see them rushing out to make movies and write books about dudes who take apart RCX modules of Lego mindstorms.
There's no conflict there, that's why. Anything illegal makes the news - especially something which most people don't know a lot about. The more exotic it is, the more likely there's a story in it. However, the media doesn't define what hacking is - it just shows one aspect of it. (Take this from someone who was a member of the media for 5 years).
I see women who love hackers.. why power. intoxicating.
Hmmm... never needed to show my "hacking prowess" to impress a woman before... Then again, I guess some people need crutches like that.
its romantic its adventure its exciting.
So is a candle-lit dinner, or white-river rafting. I don't know of many hackers who hack because it's "romantic". The adventure/excitement side of it can be found in anything, if it interests you.
it makes me piss my frigging pants!
Maybe you should try something which doesn't have such an extreme effect on your control of your bodily functions.
that you dont know shite. Why don't you go hack Janet or something lame like that? if you think emmanuel goldstein is cool. We know you are *NOT* cool. btw: i was in 8lgm! 8lgm! The best this in your liphe!
And I should care why?
Hacking is a state of mind - it doesn't mean you have to break into things to be "cool". It means that you thirst for knowledge, and you don't give up until you get it. And then you thirst for more.
It's a thirst that is never quenched.
You don't have to be part of some psuedo-cool subculture to be a hacker. If you're a loner, why identify with any group?
How do you remember it wasn't just me who was blind drunk that morning was it. I thought Emmanuel always looked kinda rough, he has his critics but in a scene with so few 'spokesmen' IMO he is as good as they get. (i was the drunk irish fella)
*grins* Mainly because I wasn't that drunk - I'd spent most of the night either chatting to Dark Tangent and co(horts:) ) about various wonderful conspiracy theories (things like microwave/shortwave-band resonance strips in $100 bills)... and spent the rest of the night chatting up a rather interesting French girl;)
BTW: I was the journo in the hat and the coat who decided it was much more fun to join in the fun than to write about it;)
... the guy from IBM doesn't know what the *hell* he's talking about. Talk about painting people with a broad brush.
As for Emmanuel Goldstein - what a star. Though he looks kinda different from when I had breakfast with him in a greasy spoon in Manchester, England after an all-night hacking run. 1996 - an excellent year;)
IIRC, Microsoft came out with a light-operated mouse similar to the one on Sun's SPARC and called it their own. Everyone sure noone has come up with this before?
Microsoft's Intellimouse Explorer doesn't require a special reflective mousepad (which becomes useless after wear and tear) like the Sun ones...
Sorry, but they're completely different implementations.
It's all wonderful advice... the other things to remember are:
Get yourself out into social situations. Become a regular at a bar. Go to parties. Act as a listener, rather than a broadcaster - you're much more interesting if you're asking people about themselves, as they're on familiar ground, and know a lot about that stuff:) It also lets them open up and gives you more chance for conversation - which is great if you're shy. People don't tend to be interested in people who aren't interested in them:)
Other things:
There's an element of risk in everything. If a girl is looking deep into your eyes, and brushing her hair back, and seems to be... expecting something, then touch her. You'll find out quickly enough if that touch was uninvited. And I don't mean touch her in a sexual way - just touch her arm, or rub her back lightly.
Basically - trust your instincts. Don't worry about being wrong too much. (Difficult for a geek, I know). Also, it's not binary. There are millions of shades of gray in real human interaction - the trick is to learn to accept that and go with it, rather than to reduce everything down to if/else clauses.:)
It's not enough just to let this guy in because he had a company, money, and some press contacts to help him. The next 17 year old they won't let in could be an open source developer without a company to his name but a damn bit more important to the computer industry.
Er... hello? COMDEX is not a place where developers get together to have a chinwag - it's a BUSINESS TO BUSINESS show.
Also - an open source developer without a company to his name isn't important to the computer industry at all - because he's operating outside of the industry. If such a developer, say, became affiliated with Red Hat, he could go into the show that way. But he'd be representing Red Hat.
Besides... it's dead easy to set up your own company - just apply for a business license. Then, who cares?
I avoid pirating software, except when it comes to Microsoft. I refuse to purchase anything they make, and I make it a point to only use their OS products. I won't pay a Microsoft tax to run the software I want to use.
They've come by here in MN, busting places that pirate Windows wholesale, and their explanation is that they're "protecting consumers". They can't even admit they're doing it to protect their bottom line. I'm a consumer, and I prefer pirated copies.
Ahhh... I understand your confusion. You see, you're not a consumer, you're actually a thief.
Probably a subtle difference in your mind, but it's an important distinction to make.
Will Gates be remembered as one of the Great Names of computer science? I like to think not. Cerf will be remembered in the future, probably more than he is now; perhaps Postel will be also; Torvalds and Stallman might; and so might Schneier (because I think cryptography is destined to become far more important in daily life than it is now). But Gates will be remembered only as a great businessman (along with Rockafeller). Or is that just me hoping?
If you're going to count Torvalds and Stallman as being members of the "Great Names of Computer Science", it's hypocritical not to include Gates. Torvalds hasn't given much to Computer Science - neither has Stallman.
In another response, you write that: "Torvalds did make a point about the efficiency of a monolithic kernel"
Hell, *I* could have told you that a monolithic kernel was efficient - it's bleeding OBVIOUS. As for "and Stallman about the possibility of writing a reasonably efficient cross-platform C compiler"... what about Ron Cain? James Hendrix? Anders Hejlsberg? To be honest, a "Cross-platform C compiler" wasn't exactly new - people had been writing them for years, and more efficiently than Stallman (see HiSoft, for example).
Tellya what... include Gates, and I'll let you include Torvalds and Stallman.
This remind me of a review of Ofice 2000 in PC World (India). The review made the following relevant points: 1> Office 2000 saves html documents(save as web page or some such option) in a broken XML format, support for which is available only in IE 5. 2> The page cannot be displayed at all w/ netscape.
If M$ plans to use the same format for SOAP...
Funny... the pages display perfectly fine in Opera. It's Netscape's crappy CSS/XML/XSL/DHTML implementation.
Simon
Re:Geem, where have I seen this before...
on
Time Doesn't Exist
·
· Score: 1
Speaking of papers, I've got one online that gives a geometrical argument for the absence of time as a factor in particle measurements (that is, measurements affect the particle's entire history and future).
http://home.earthlink.com/~simoncooke/phys ics
I've been playing with this idea for a few years now.. but this is the first time I've published it;)
What really amuses me is everyone saying that "Linux never crashes" and "Linux is stable", and claiming that NT crashes all the time, when in fact it's the same scenario as when the NT GUI goes down; the kernel is still going, it's just that you can't do anything from the GUI.
The only way to specify externally where to look for a DLL is PATH. PATH is also used for finding executables. Weird to use it for both. A user may have to set some directory in front of it's path but not want to have the DLL's in that dir.
Even more stupid is that PATH is only searched after all other possibilities.
There should be a unique environment variable for this, which overrides other mechanisms such as \WINDOWS etc.
Why not just put a symbolic link in the application directory to the DLL you want to specify?
You claim it to be simple, yet use nearly 2 pages of HTML to describe the process to find a DLL. And it doesn't even cover incompatible versions of DLLs, which is a PAIN in NT.
Sorry for being verbose; here's the short version (standard search mechanism only):
The directory from which the application loaded.
The current directory.
Windows 95/98: The Windows system directory. Use the GetSystemDirectory function to get the path of this directory.
Windows NT/ 2000: The 32-bit Windows system directory. Use the GetSystemDirectory function to get the path of this directory. The name of this directory is SYSTEM32.
Windows NT/ 2000: The 16-bit Windows system directory. There is no function that obtains the path of this directory, but it is searched. The name of this directory is SYSTEM.
The Windows directory. Use the GetWindowsDirectory function to get the path of this directory.
The directories that are listed in the PATH environment variable.
As for "incompatible DLLs", if an app requires a specific version of a DLL, make sure it's in the app's directory, or put a link to the correct DLL version in that directory (requires NT running on NTFS for this).
And you people blaming the apps for all these problems are just plain wrong. If Microsoft had provided a decent Installer, simple methods of finding DLLs+other components and a structured way of storing them, the clueless application makers of NT would have done the right thing. Do you really expect every company out there to make their own installer?
Microsoft publishes guidelines about how to do it right. App vendors generally don't bother with them, but are getting better. If they *DID* bother with them, you wouldn't have all these problems.
If someone ignores the standard linux way of doing things, you're going to have similar problems.
Why do applications clutter their DLLs all over the SYSTEM directory? Is this the application makers fault too? Do you really expect every individual company to do the right thing, the same thing and stay compatible. Not following the leader's example?
Yes, it's the application makers fault. There's NO reason to stick DLLs in the System directory unless you're Microsoft. End Of Story. It's completely avoidable, and is even recommended *AGAINST*. People still do it, however. Why is a mystery. Just goes to show that there are a lot of coders out there who don't know what they're doing.
You conviniently left out the hassles of OCXes and other things which must be installed into the registry. (Another chapter all by itself)
What? You mean OCXs and other things that must be installed into the registry, like COM components? You mean objects that for any given version have a specific UUID for their object, which means that the whole versioning issue should be moot?
How can anybody in their right mind defend the DLL "system"?
How about because it's nigh-on identical to the Linux system, but most end-users of Windows systems aren't able to use symbolic links to get around the hassles because Win98/95 doesn't support them? Not to mention the fact that most end-users of Windows systems aren't educated in the technology to the point where it would be advantageous to them to control where DLLs are installed, manage the symlink map, etc etc etc?
And I'm NOT saying the Linux system for libraries is perfect, it's just better handled by competent developers and semi-standard methods. It could always be made better, especially with restrictions and structure regulations from the system. Or just an agreement of standards among Linux distributions. It's incredible how little developers fuck up the system when they have the right tools for the job.
Actually, it sounds like it's nothing to do with having "the right tools for the job" - you've just said it requires competent developers and semi-standard methods. There's nothing special about the Linux way of doing it - it's just that people heed the warning about the right way to do things more than Windows developers have done in the past.
Microsoft has standards for doing this kind of thing. It's just that people have a strange ability to ignore them at times.
I'm a Windows developer. I know that the only things that go in the System directory are: Microsoft system libraries and system components. Device drivers provided by Microsoft. Windows configuration files and data (provided by Microsoft). Kernel files.
I'm hedging my bets it will be fish shaped, and will fit into the inner ear.
Well, if it is, I think it died at some point while it was stuck in someone's lughole...
Reason being? Well... reading the English info about the project (which I can only assume was run through their "enconverter" and "deconverter"):
Multi-lingual network aims to enable people to communicate in their mother language with peoples of different language. UNL is a common language shared by people over the world in multi-lingual network. UNL system basically consists of network and conversion program between UNL and native languages. A conversion system from native languages into UNL is called "enconverter", and that from UNL into native languages is called "deconverter". Information in each language, being "enconverted", is exchanged via network in the form of UNL. Information represented in UNL is "deconverted" into each native language on the terminal of network. In transmission of information, the information which is expressed in a mother language is enconverted into UNL. Preciseness of conversion can be verified by deconverting the UNL representation into the language from which the UNL is obtained.
Why not call it "encoding" and "decoding" like the rest of the world?
Internet is spelt "Internet", not "Inter-Net"
The grammar is terrible. "Information represented in UNL is "deconverted" into each native language on the terminal of network."... er... anyone else see words missing from that sentence?
I'm just hoping that they get people who can actually read and write their target languages fluently to do the testing...
Slashdot Mirror
I wrote:
Same end result, two different mechanisms. One of which you favor.
Then Eric wrote:
Building added functionality into the operating system is not the issue. The issue is Microsoft coming out with their own version of a proxy server and then threatening Dell that they are going to withold their Windows license if they don't preload their server and only their server along with Windows. Or if they call up Apple and say "we'll cancel Mac Office if you don't preload our Mac proxy server, and deep six your own version for Windows."
But what does this have to do with the post I responded to? Answer: absolutely nothing.
Hmmm... you would appear to be correct. It's probably all that OLE structured storage stuff that's ballooning it up.
Take internet connection sharing for example. I've gone through wingate, sygate, and finally winroute, all shareware applications operating in a viable market. Now microsoft in windows 98 second edition has come out with 'internet connection sharing' - which makes the entire market defunct!!! Microsoft is going around squashing honest programmers and companies.
So what would you do if I'd ported the same functionality from Linux over to Windows and released it under the GPL? Probably nothing. You wouldn't be jumping up and down about it.
Same end result, two different mechanisms. One of which you favor.
Hypocrit.
Word does, after all, store the file as UNICODE. So it'll be ~ twice as long.
"For while consumers might one day turn to network
computers, or Linux, or a combination of middleware and some other operating system, as an
alternative to Windows, the fact remains that they are not doing so today. Nor are consumers
likely to do so in appreciable numbers any time in the next few years." - Judge Jackson.
Welcome to the ISO-standardized mechanism for file handling.
1. Create temporary filename.
2. Create temporary file.
3. Copy entire file into another temporary file.
4. Dance on that file for a while.
5. Delete Original File.
6. Rename Temp file to Original filename.
7. Touch the file to update the file modification date and creation dates to match the original file.
Failure to use this mechanism will result in a slow and painful death, as administered by Nick, because as we all know, in computing there's only one TRUE way of doing anything, and anything other than the way you know and understand is obviously (1) inferior, (2) error prone, and (3) the spawn of Satan.
Pending Standards:
Mouse button click duration; accurate to the nearest millisecond. Button-clicks by users outside of this 1ms time window will be ignored.
Standard function names. All function names must begin with the word "Fondue" for no apparent reason, other than because when I started programming 9 years ago, it seemed like a good idea at the time.
All programs must be written in C. No other languages are allowed.
When performing large file updates, you must create a delta file, which you then run through and create a "child" file from, by comparing the delta file to the parent file. You then must backup the parent file. You must perform all actions as three separate processes: add new records, sort, remove duplicates. Preferably overnight, when most people won't be using the mainframe time. (Oh sorry - home users don't have mainframes? Well, tough titty).
I became aware of another potentially anti-competitive thing MS did some years back, which, as far as I am aware no one else has mentioned in the press before. If it wasn't anti-competitive, it was at least stupid, and MS users are the ultimate victims.
You should get out more. You'll be telling me that it's anticompetitive to support HTML 4.0 in web documents soon, because "Mosaic 1.0 doesn't support it".
Amen! The man is obviously not a hacker. He cites blue boxing as his principal hacking experience, something that's been impossible for over a decade.
That's so amusing - it was still possible in 1994.
Simon
umm.. i dont see them rushing out to make movies and write books about dudes who take apart RCX modules of Lego mindstorms.
There's no conflict there, that's why. Anything illegal makes the news - especially something which most people don't know a lot about. The more exotic it is, the more likely there's a story in it. However, the media doesn't define what hacking is - it just shows one aspect of it. (Take this from someone who was a member of the media for 5 years).
I see women who love hackers.. why power. intoxicating.
Hmmm... never needed to show my "hacking prowess" to impress a woman before... Then again, I guess some people need crutches like that.
its romantic
its adventure
its exciting.
So is a candle-lit dinner, or white-river rafting. I don't know of many hackers who hack because it's "romantic". The adventure/excitement side of it can be found in anything, if it interests you.
it makes me piss my frigging pants!
Maybe you should try something which doesn't have such an extreme effect on your control of your bodily functions.
Simon
that you dont know shite. Why don't you go hack Janet or something lame like that? if you think emmanuel goldstein is cool. We know you are *NOT* cool. btw: i was in 8lgm! 8lgm! The best this in your liphe!
And I should care why?
Hacking is a state of mind - it doesn't mean you have to break into things to be "cool". It means that you thirst for knowledge, and you don't give up until you get it. And then you thirst for more.
It's a thirst that is never quenched.
You don't have to be part of some psuedo-cool subculture to be a hacker. If you're a loner, why identify with any group?
Simon
How do you remember it wasn't just me who was blind drunk that morning was it. I thought Emmanuel always looked kinda rough, he has his critics but in a scene with so few 'spokesmen' IMO he is as good as they get. (i was the drunk irish fella)
:) ) about various wonderful conspiracy theories (things like microwave/shortwave-band resonance strips in $100 bills)... and spent the rest of the night chatting up a rather interesting French girl ;)
;)
*grins* Mainly because I wasn't that drunk - I'd spent most of the night either chatting to Dark Tangent and co(horts
BTW: I was the journo in the hat and the coat who decided it was much more fun to join in the fun than to write about it
Simon
Damnit... my <NAMEDROP;> tags didn't come out right...
... the guy from IBM doesn't know what the *hell* he's talking about. Talk about painting people with a broad brush.
;)
As for Emmanuel Goldstein - what a star. Though he looks kinda different from when I had breakfast with him in a greasy spoon in Manchester, England after an all-night hacking run. 1996 - an excellent year
:-)
Simon
A little knowledge is a dangerous thing
IIRC, Microsoft came out with a light-operated mouse similar to the one on Sun's SPARC and called it their own. Everyone sure noone has come up with this before?
Microsoft's Intellimouse Explorer doesn't require a special reflective mousepad (which becomes useless after wear and tear) like the Sun ones...
Sorry, but they're completely different implementations.
Simon
A little knowledge is a dangerous thing
It's all wonderful advice... the other things to remember are:
:) It also lets them open up and gives you more chance for conversation - which is great if you're shy. People don't tend to be interested in people who aren't interested in them :)
:)
Get yourself out into social situations. Become a regular at a bar. Go to parties. Act as a listener, rather than a broadcaster - you're much more interesting if you're asking people about themselves, as they're on familiar ground, and know a lot about that stuff
Other things:
There's an element of risk in everything. If a girl is looking deep into your eyes, and brushing her hair back, and seems to be... expecting something, then touch her. You'll find out quickly enough if that touch was uninvited. And I don't mean touch her in a sexual way - just touch her arm, or rub her back lightly.
Basically - trust your instincts. Don't worry about being wrong too much. (Difficult for a geek, I know). Also, it's not binary. There are millions of shades of gray in real human interaction - the trick is to learn to accept that and go with it, rather than to reduce everything down to if/else clauses.
Simon
It's not enough just to let this guy in because he had a company, money, and some press contacts to help him. The next 17 year old they won't let in could be an open source developer without a company to his name but a damn bit more important to the computer industry.
Er... hello? COMDEX is not a place where developers get together to have a chinwag - it's a BUSINESS TO BUSINESS show.
Also - an open source developer without a company to his name isn't important to the computer industry at all - because he's operating outside of the industry. If such a developer, say, became affiliated with Red Hat, he could go into the show that way. But he'd be representing Red Hat.
Besides... it's dead easy to set up your own company - just apply for a business license. Then, who cares?
Simon
I avoid pirating software, except when it comes to Microsoft. I refuse to purchase anything they make, and I make it a point to only use their OS products. I won't pay a Microsoft tax to run the software I want to use.
They've come by here in MN, busting places that pirate Windows wholesale, and their explanation is that they're "protecting consumers". They can't even admit they're doing it to protect their bottom line. I'm a consumer, and I prefer pirated copies.
Ahhh... I understand your confusion. You see, you're not a consumer, you're actually a thief.
Probably a subtle difference in your mind, but it's an important distinction to make.
Simon
Wow. After reading this, I wish there was an "Out of his gourd" option on the moderation box.
Will Gates be remembered as one of the Great Names of computer science? I like to think not. Cerf will be remembered in the future, probably more than he is now; perhaps Postel will be also; Torvalds and Stallman might; and so might Schneier (because I think cryptography is destined to become far more important in daily life than it is now). But Gates will be remembered only as a great businessman (along with Rockafeller). Or is that just me hoping?
If you're going to count Torvalds and Stallman as being members of the "Great Names of Computer Science", it's hypocritical not to include Gates. Torvalds hasn't given much to Computer Science - neither has Stallman.
In another response, you write that:
"Torvalds did make a point about the efficiency of a monolithic kernel"
Hell, *I* could have told you that a monolithic kernel was efficient - it's bleeding OBVIOUS. As for "and Stallman about the possibility of writing a reasonably efficient cross-platform C compiler"... what about Ron Cain? James Hendrix? Anders Hejlsberg? To be honest, a "Cross-platform C compiler" wasn't exactly new - people had been writing them for years, and more efficiently than Stallman (see HiSoft, for example).
Tellya what... include Gates, and I'll let you include Torvalds and Stallman.
Simon
This remind me of a review of Ofice 2000 in PC World (India). The review made the following relevant points:
1> Office 2000 saves html documents(save as web page or some such option) in a broken XML format, support for which is available only in IE 5.
2> The page cannot be displayed at all w/ netscape.
If M$ plans to use the same format for SOAP...
Funny... the pages display perfectly fine in Opera. It's Netscape's crappy CSS/XML/XSL/DHTML implementation.
Simon
Speaking of papers, I've got one online that gives a geometrical argument for the absence of time as a factor in particle measurements (that is, measurements affect the particle's entire history and future).
;)
http://home.earthlink.com/~simoncooke/phys ics
I've been playing with this idea for a few years now.. but this is the first time I've published it
Simon
What really amuses me is everyone saying that "Linux never crashes" and "Linux is stable", and claiming that NT crashes all the time, when in fact it's the same scenario as when the NT GUI goes down; the kernel is still going, it's just that you can't do anything from the GUI.
Simon
LoadLibraryEx is stupid, since:
The only way to specify externally where to look for a DLL is PATH. PATH is also used for finding executables. Weird to use it for both. A user may have to set some directory in front of it's path but not want to have the DLL's in that dir.
Even more stupid is that PATH is only searched after all other possibilities.
There should be a unique environment variable for this, which overrides other mechanisms such as \WINDOWS etc.
Why not just put a symbolic link in the application directory to the DLL you want to specify?
Simon
Sorry for being verbose; here's the short version (standard search mechanism only):
As for "incompatible DLLs", if an app requires a specific version of a DLL, make sure it's in the app's directory, or put a link to the correct DLL version in that directory (requires NT running on NTFS for this).
And you people blaming the apps for all these problems are just plain wrong. If Microsoft had provided a decent Installer, simple methods of finding DLLs+other components and a structured way of storing them, the clueless application makers of NT would have done the right thing. Do you really expect every company out there to make their own installer?
Microsoft publishes guidelines about how to do it right. App vendors generally don't bother with them, but are getting better. If they *DID* bother with them, you wouldn't have all these problems.
If someone ignores the standard linux way of doing things, you're going to have similar problems.
Why do applications clutter their DLLs all over the SYSTEM directory? Is this the application makers fault too? Do you really expect every individual company to do the right thing, the same thing and stay compatible. Not following the leader's example?
Yes, it's the application makers fault. There's NO reason to stick DLLs in the System directory unless you're Microsoft. End Of Story. It's completely avoidable, and is even recommended *AGAINST*. People still do it, however. Why is a mystery. Just goes to show that there are a lot of coders out there who don't know what they're doing.
You conviniently left out the hassles of OCXes and other things which must be installed into the registry. (Another chapter all by itself)
What? You mean OCXs and other things that must be installed into the registry, like COM components? You mean objects that for any given version have a specific UUID for their object, which means that the whole versioning issue should be moot?
How can anybody in their right mind defend the DLL "system"?
How about because it's nigh-on identical to the Linux system, but most end-users of Windows systems aren't able to use symbolic links to get around the hassles because Win98/95 doesn't support them? Not to mention the fact that most end-users of Windows systems aren't educated in the technology to the point where it would be advantageous to them to control where DLLs are installed, manage the symlink map, etc etc etc?
And I'm NOT saying the Linux system for libraries is perfect, it's just better handled by competent developers and semi-standard methods. It could always be made better, especially with restrictions and structure regulations from the system. Or just an agreement of standards among Linux distributions. It's incredible how little developers fuck up the system when they have the right tools for the job.
Actually, it sounds like it's nothing to do with having "the right tools for the job" - you've just said it requires competent developers and semi-standard methods. There's nothing special about the Linux way of doing it - it's just that people heed the warning about the right way to do things more than Windows developers have done in the past.
Microsoft has standards for doing this kind of thing. It's just that people have a strange ability to ignore them at times.
I'm a Windows developer. I know that the only things that go in the System directory are:
Microsoft system libraries and system components.
Device drivers provided by Microsoft.
Windows configuration files and data (provided by Microsoft).
Kernel files.
NOT User Application files.
Simon
Babelfish...right?
:)
Unless the joke was the whole UNL site? Which I can see
Si
Well, if it is, I think it died at some point while it was stuck in someone's lughole...
Reason being? Well... reading the English info about the project (which I can only assume was run through their "enconverter" and "deconverter"):
Multi-lingual network aims to enable people to communicate in their mother language with peoples of different language. UNL is a common language shared by people over the world in multi-lingual network. UNL system basically consists of network and conversion program between UNL and native languages.
A conversion system from native languages into UNL is called "enconverter", and that from UNL into native languages is called "deconverter". Information in each language, being "enconverted", is exchanged via network in the form of UNL. Information represented in UNL is "deconverted" into each native language on the terminal of network.
In transmission of information, the information which is expressed in a mother language is enconverted into UNL. Preciseness of conversion can be verified by deconverting the UNL representation into the language from which the UNL is obtained.
I'm just hoping that they get people who can actually read and write their target languages fluently to do the testing...
Simon