AFAIR the windows theading model is better, because windows was designed with heavily multithreading in mind, and a very expensive operation of creation of new processes.
In Linux is the other way around: forking a process is very efficient, so no one used threads so they were not as efficient as the ones in Windows.
Of course both process creation in Windows and threading in Linux could have been perfected in all this years, but that is something I don't know.
That said, a bug in a thread will bring the application or server down, and a bug in a forked process is trivial to correct: just restart the process.
Well, I've been thinking that for several years Miguel de Icaza has being in MS payroll.
First, Linux should stop having an unified API towards which to port software. MS knows the value of an API, and that's the reason they have kept backwards compatibility between win95 to XP and Vista. Disrupt the API. So Gnome was born. Should I code for KDE.. or Gnome?
Then the.Mono stuff, open Linux distros to patent infringement.
But it was not enough, so now Novell has made an agreement with MS. And Icaza should be behind that too. I don't think any of it is a coincidence.
I understand and respect your preference for VB syntax.
However I vastly prefer C++ syntax (except for templates, wich I don't use). Why? because VB6 simply didn't let me do things I wanted to do. I felt extremely frustrated. I have not used VB.Net, so I can't comment on it.
I can catch those (and more complex) errors just by inspecting the code. So your example is simply silly to me.
Because I know how to make parsers to process it (wich are slower than parsers for about any other form of data).
Because I know s-expressions which are easier to type and encode much more information that XML. s-expressions can be data exactly like XML, but can also be code. Beautiful code.
Because is more a marketing buzzword than real meat.
That said, I use XML when I have to, like generating XLS files in PHP. But I would not use it at all if I didn't received money for doing it.
I am 100% physicalist and yet I value life over everything else. In fact I tend to value everything I can't fully understand. Knowledge is valuable. Even if I would fully understand life, the fact that something can have consciousness is absolutely amazing and deserve the uttermost respect.
I don't see how anyone can see living things no different from non-living things. Absurd. That's just stupid FUD created two centuries ago.
However, I would not have highlighted the sentence the GP did. To me the insightful sentence in your post is: A society that embraces spirituality above all other things is hardly removed from barbarism.
Actually is both of them, but it's to me more a book than a website. A book is defined by its contents, the sequential flow of text where you keep reading it from start to end. With coherent writing and good spelling.
So I think that a book that uses HTML and CSS is still called a book. An online book, but still a book.
The GP is right. With just stereo headphones and a good sound card or source audio anyone can sense all those directions. With just two ears and two sound channels.
That's because we feel those directions by perceiving small differences in time and strength between both ears, we can't move our ears like dogs or something like that.
I still remember the matchbox demo, with that little thing going over my head, behind me, under me, and so on.
With crappy speakers, you will only hear front sound, that's a given point.
I have both a surround system and a good pair of headphones and I can perceive the directions with both of them (however the surround system makes it more real, because you can perceive low frequency vibrations with more parts of your body than just your ears).
In C++, some memory leaks can only be hunted and eliminated with an interactive debugger, so if that's what you were talking about, you're right.
But I don't see it necessary at all in an interpreted language, and 99% of code nowadays is written in interpreted or managed languages.
You can use printf, echo, Response.Write(), wxMessageBox() or whatever to debug a program without having to use an interactive debugger at all. I made a windows DLL recently, and I know I didn't used a debugger. I just used the return value of the exported function in the DLL to do all the debugging.
Knuth in his AoCP series has a paragraph or two about that.
In essence, only infinite sequences can be truly random.
However, we can define a "randomness" measure for finite sequences, and it has to do with the distribution of the digits and some other stuff I don't remember at the moment.
So the GP really has an argument, even if he misses some details.
My point is, you need to read about "randomness" as much as he. I do need to re-read about it too.
I like the general idea. In C you have the "static" keywork that makes a variable keep their values between calls of a function. I would not propose an "always persistent" memory, because too many calculations are temporary, but a "persistent" keyword in the most used programming language would be a very nice thing. However, there is the entire databases thing, they use another paradigm that should be taken into account.
I watch both TV and DVDs in a 19" CRT display, and the DVDs look much better. You can also buy one of those expensive HD TVs.
When I see a normal TV all I can think is "that screen looks way too fuzzy/blurred".
Re:Implementations (lack of)
on
Lisp and Ruby
·
· Score: 1
I use Slime. I like some features of it a lot.
But I don't like the fact that the history of s-expressions behaves so akwardly after I became used to the way that history of inputs behaves in chat programs like mirc. I can't edit or select something and keep browsing the history, I have to restart. Also selecting and moving text with the mouse is awful. It's so easy with ALL other programs (except Word or Office, of course), that its at least unsettling. You can really feel that it was designed for keyboard-only users and that mouse support was an afterthough.
Maybe is not Emacs but Slime, but anyway, what else can you use for Lisp?
Re:Performance, anyone?
on
Lisp and Ruby
·
· Score: 1
I was using Python as "the slowest of mainstream computer languages" for comparison with Ruby. In fact that's the point I was trying to make.
So you're telling me that Python is faster than PHP? That's news to me.
Re:Implementations (lack of)
on
Lisp and Ruby
·
· Score: 1
And of course, if I would need to write code that only used the Common Lisp Spec, then yes, you would be totally right.
But somehow I always need to do more esoteric stuff.
Re:Implementations (lack of)
on
Lisp and Ruby
·
· Score: 1
You're right in the general sense.
As other poster said, CLISP is the closest (and only one) Lisp implementation that is even close to my requirements. But if you need to do something outside the Common Lisp Standard in CLISP, you need to use CLISP extensions to the standard, or modify CLISP. And if you do that your Lisp code will be dependent on CLISP and if it's dependent on CLISP the CLISP mantainers will claim that your code will need to be under the GPL.
As the Common Lisp Standard is very old and lacks Threads and some other very useful things like a GUI, is not hard to find yourself in that position, like what happened once with wxCL.
I know UFFI and other libraries are working on trying to fix that. But right now, it's not there.
To me this is more like: Microsoft has now proved the most popular reason to dump Microsoft DRM: the Zune player forces users to rebuy their legit content, because its MS DRM conflicts with the old MS DRM.
Let MS enjoy their Desktop OS monopoly, but never, ever, give them any other monopoly to control us.
I would like that Daemon tools would emulate a CD/DVD burner too. Then no CDs would be wasted in the process.
Re:Performance, anyone?
on
Lisp and Ruby
·
· Score: 1
Ruby is even slower than Phyton.
It's simply that a lot of though has been put into those Lisp compilers as they used to run in very small computers compared with the ones we have.
Implementations (lack of)
on
Lisp and Ruby
·
· Score: 0, Troll
There is not a good Lisp free implementation that runs on several platforms including Windows and that not makes all the code I write there GPL instead of the license I would like.
Having said that, I'm really waiting for news of SBCL running in Windows with threading and all.
However, that's not enough, I would like to have Threads standarised in Lisp and also a GUI for Lisp that's like wxWidgets for C++.
Also, Emacs sucks, and it's the best IDE there is for Lisp. I know some people are mouse-impaired and live for and by the keyboard, but we Starcraft players know better, we must use both very fast. (I have found myself using the odd and hand twisting keyboard commands of Emacs in more sane software, and I was shocked in awe.)
So in short: I would like to use Lisp for a lot of projects, but I can't without investing USD$2000+ (and I can't do it for now). And I would like a better IDE.
Compare that with PHP, Perl, Phyton, Ruby, Java and C++: All of them have non-GPL infecting complete implementations that are available for Windows.
Notes: I tried wxCL but it didn't compiled out of the box in my mingw instalation. I couldn't make it run at all. ECL is LGPL but as the LGPL requires that code linked statically has also to be under the LGPL, then ECL has a license as infectious as the GPL.
Slashdot Mirror
I did the same, but I had to use forced encryption for it to work, otherwise they would cripple all the unencrypted traffic.
AFAIR the windows theading model is better, because windows was designed with heavily multithreading in mind, and a very expensive operation of creation of new processes.
In Linux is the other way around: forking a process is very efficient, so no one used threads so they were not as efficient as the ones in Windows.
Of course both process creation in Windows and threading in Linux could have been perfected in all this years, but that is something I don't know.
That said, a bug in a thread will bring the application or server down, and a bug in a forked process is trivial to correct: just restart the process.
You can't create documents in word with good styling.
But if you did it in WP51, and then converted it to Word, it worked perfectly fine.
So I guess the Word engine can handle them just right, but the interface is incapable of generating them the right way.
Well, I've been thinking that for several years Miguel de Icaza has being in MS payroll.
.Mono stuff, open Linux distros to patent infringement.
First, Linux should stop having an unified API towards which to port software. MS knows the value of an API, and that's the reason they have kept backwards compatibility between win95 to XP and Vista. Disrupt the API. So Gnome was born. Should I code for KDE.. or Gnome?
Then the
But it was not enough, so now Novell has made an agreement with MS. And Icaza should be behind that too. I don't think any of it is a coincidence.
I understand and respect your preference for VB syntax.
However I vastly prefer C++ syntax (except for templates, wich I don't use). Why? because VB6 simply didn't let me do things I wanted to do. I felt extremely frustrated. I have not used VB.Net, so I can't comment on it.
I can catch those (and more complex) errors just by inspecting the code. So your example is simply silly to me.
My point? To each its own.
... counseling business, of course.
- the-code
http://geekz.co.uk/lovesraymond/archive/show-them
Because I know how to make parsers to process it (wich are slower than parsers for about any other form of data).
Because I know s-expressions which are easier to type and encode much more information that XML. s-expressions can be data exactly like XML, but can also be code. Beautiful code.
Because is more a marketing buzzword than real meat.
That said, I use XML when I have to, like generating XLS files in PHP. But I would not use it at all if I didn't received money for doing it.
I am 100% physicalist and yet I value life over everything else. In fact I tend to value everything I can't fully understand. Knowledge is valuable. Even if I would fully understand life, the fact that something can have consciousness is absolutely amazing and deserve the uttermost respect.
I don't see how anyone can see living things no different from non-living things. Absurd. That's just stupid FUD created two centuries ago.
However, I would not have highlighted the sentence the GP did. To me the insightful sentence in your post is: A society that embraces spirituality above all other things is hardly removed from barbarism.
Priceless.
It looks you would enjoy working with LaTeX.
You can specify everything to the smallest detail.
I like the syntax, however some people can't get used to it. YMMV.
http://mitpress.mit.edu/sicp/full-text/book/book.h tml
Is this a book? or a Website?
Actually is both of them, but it's to me more a book than a website. A book is defined by its contents, the sequential flow of text where you keep reading it from start to end. With coherent writing and good spelling.
So I think that a book that uses HTML and CSS is still called a book. An online book, but still a book.
The GP is right. With just stereo headphones and a good sound card or source audio anyone can sense all those directions. With just two ears and two sound channels.
That's because we feel those directions by perceiving small differences in time and strength between both ears, we can't move our ears like dogs or something like that.
I still remember the matchbox demo, with that little thing going over my head, behind me, under me, and so on.
With crappy speakers, you will only hear front sound, that's a given point.
I have both a surround system and a good pair of headphones and I can perceive the directions with both of them (however the surround system makes it more real, because you can perceive low frequency vibrations with more parts of your body than just your ears).
In C++, some memory leaks can only be hunted and eliminated with an interactive debugger, so if that's what you were talking about, you're right.
But I don't see it necessary at all in an interpreted language, and 99% of code nowadays is written in interpreted or managed languages.
You can use printf, echo, Response.Write(), wxMessageBox() or whatever to debug a program without having to use an interactive debugger at all. I made a windows DLL recently, and I know I didn't used a debugger. I just used the return value of the exported function in the DLL to do all the debugging.
Just play any PC game that works in a LAN.
You play Online, and you play with your friends.
If you want to chat with them, install Ventrilo too.
Knuth in his AoCP series has a paragraph or two about that.
In essence, only infinite sequences can be truly random.
However, we can define a "randomness" measure for finite sequences, and it has to do with the distribution of the digits and some other stuff I don't remember at the moment.
So the GP really has an argument, even if he misses some details.
My point is, you need to read about "randomness" as much as he. I do need to re-read about it too.
I have to add, that this "labor" doesn't apply to everybody.
You have to be already smart and creative to begin with. More creative than smart, which is harder to find.
I like the general idea. In C you have the "static" keywork that makes a variable keep their values between calls of a function.
I would not propose an "always persistent" memory, because too many calculations are temporary, but a "persistent" keyword in the most used programming language would be a very nice thing.
However, there is the entire databases thing, they use another paradigm that should be taken into account.
If you watch them in a normal TV, then yes.
I watch both TV and DVDs in a 19" CRT display, and the DVDs look much better. You can also buy one of those expensive HD TVs.
When I see a normal TV all I can think is "that screen looks way too fuzzy/blurred".
I use Slime. I like some features of it a lot.
But I don't like the fact that the history of s-expressions behaves so akwardly after I became used to the way that history of inputs behaves in chat programs like mirc. I can't edit or select something and keep browsing the history, I have to restart. Also selecting and moving text with the mouse is awful. It's so easy with ALL other programs (except Word or Office, of course), that its at least unsettling. You can really feel that it was designed for keyboard-only users and that mouse support was an afterthough.
Maybe is not Emacs but Slime, but anyway, what else can you use for Lisp?
I was using Python as "the slowest of mainstream computer languages" for comparison with Ruby. In fact that's the point I was trying to make.
So you're telling me that Python is faster than PHP? That's news to me.
And of course, if I would need to write code that only used the Common Lisp Spec, then yes, you would be totally right.
But somehow I always need to do more esoteric stuff.
You're right in the general sense.
As other poster said, CLISP is the closest (and only one) Lisp implementation that is even close to my requirements. But if you need to do something outside the Common Lisp Standard in CLISP, you need to use CLISP extensions to the standard, or modify CLISP.
And if you do that your Lisp code will be dependent on CLISP and if it's dependent on CLISP the CLISP mantainers will claim that your code will need to be under the GPL.
As the Common Lisp Standard is very old and lacks Threads and some other very useful things like a GUI, is not hard to find yourself in that position, like what happened once with wxCL.
I know UFFI and other libraries are working on trying to fix that. But right now, it's not there.
To me this is more like:
Microsoft has now proved the most popular reason to dump Microsoft DRM: the Zune player forces users to rebuy their legit content, because its MS DRM conflicts with the old MS DRM.
Let MS enjoy their Desktop OS monopoly, but never, ever, give them any other monopoly to control us.
Apple does DRM right. At least to me.
I would like that Daemon tools would emulate a CD/DVD burner too. Then no CDs would be wasted in the process.
Ruby is even slower than Phyton.
It's simply that a lot of though has been put into those Lisp compilers as they used to run in very small computers compared with the ones we have.
There is not a good Lisp free implementation that runs on several platforms including Windows and that not makes all the code I write there GPL instead of the license I would like.
Having said that, I'm really waiting for news of SBCL running in Windows with threading and all.
However, that's not enough, I would like to have Threads standarised in Lisp and also a GUI for Lisp that's like wxWidgets for C++.
Also, Emacs sucks, and it's the best IDE there is for Lisp. I know some people are mouse-impaired and live for and by the keyboard, but we Starcraft players know better, we must use both very fast. (I have found myself using the odd and hand twisting keyboard commands of Emacs in more sane software, and I was shocked in awe.)
So in short: I would like to use Lisp for a lot of projects, but I can't without investing USD$2000+ (and I can't do it for now). And I would like a better IDE.
Compare that with PHP, Perl, Phyton, Ruby, Java and C++: All of them have non-GPL infecting complete implementations that are available for Windows.
Notes: I tried wxCL but it didn't compiled out of the box in my mingw instalation. I couldn't make it run at all. ECL is LGPL but as the LGPL requires that code linked statically has also to be under the LGPL, then ECL has a license as infectious as the GPL.