Well, I noticed my CDMA phone doens't do that and I also noticed the interference is "loudest" when I bring the phone close to the computer case. But you are probably right, I really don't know that much about electronics...
How about someone that graduated from a 4 year (1 year co-op=5 year) US university with the Computer Science degree and a 3.94 GPA? Oh, and I am a foreigner -- I just happen to have a green card?
What good is my "super Kewl" GSM phone with the SIM card that I can move to another "super kewl" GSM phone if the reception is crap. Yeah I had one of those new and flashy GSM phones except that I got bad reception both at home and at in the building were I work. I could take pictures, connect with bluetooth, play games and all but I couldn't talk on the phone.
Then switched to Verizon and got a free basic LG VX3200 and now I have a _phone_ that works as a phone. The bottom line is I am not buying a phone only - I am buying a phone with a service (network that comes with it). Verizon so far is the best in my area, everyone I know that has either Cingular or T-Mobile has crappy reception.
Also, what is up with GSM phones that they interfere with the electronics in my computer - when the phone was on my desk it would periodically induce this clicking noise in my speakers...
The more people they scare the better. This is terrorism at its best. It is not that they will be able to sue everyone who ever accessed a P2P network, but they sure can terrorize people like with just a couple of "examples" like this and then also do batch litigation a 1000 or more people at a time here and there -- just enough to have everyone's "friend of friend" go through this. When stuff happens to "people you know" then one really starts thinking that it might happen him/her.
I still consider the best strategy if one gets caught is to infect your own machine with viruses/worms/spyware/trojans - anything that will provide a possible explanation of how the files ended up on there
In my post, the topic was the resistance of harmful micro-organisms to antibiotics, which includes bacteria and viruses. What is the problem with that?
If you want to talk about viruses only, reply to a different comment and use the word "evolution" as many times as you want in your post. Nobody is going to stop you (aside from almighty moderators, of course;)
For unclear reasons you are desperate to have the word "evolution" some place in the comment so here you go: EVOLUTION, better now?
Quoting Wikipedia is the same as quoting yourself, since you could have technically gone and edited the entry to support your point of view.
You are right, Ruby is cleaner and more consistent. But Python works for me because it has more 3rd party mechanisms to inline C code and thus I can optimize say 3 nested for loops in a certain function. Ruby has been popular in Japan and Asia for quite a while but I think there is more documentaion and examples of code for Python.
I don't think it is considered evolution in the classic sense. Just because you really, really want to use the word "to stick it" to those crazy people from Kansas doesn't mean that you should. What happens is that bacteria has a specific set of genes that will mutate faster and those genes will protect it from certain molecules. So we have seen staph mutate to protect itself against different antibiotics and that can happen very quickly but staph will probably not turn into a fungus or an insect just as quickly. It seems the bacteria have evolved this fast mutating mechanism of quick adaptability to defend itself agianst toxins but it is limited to that. But the ability to have this kind of defense mechanism would be evolution in the classic sense.
I don't have enough evidence to support that, but it makes sense and others have recommended not to use antibacterial soaps at home. Here is an excerpt from some By Dr. Joseph Mercola:
In the study, published in the March 2, 2004 journal Annals of Internal Medicine, people who used antibacterial soaps and cleansers developed cough, runny nose, sore throat, fever, vomiting, diarrhea and other symptoms just as often as people who used products that did not contain antibacterial ingredients.
and then from the same article
Even the American Medical Association (AMA) does not recommend these products. So why do they persist? Simple; the manufacturers have relied on using fear to convince people that they need to use them to stay healthy.
The point is that antibacterial soap should have its uses and is recommended to be used in the hospitals. But when it is used at home and everywhere eventually the bacteria will develop resistance., because 1) A lot more bacteria is exposed to it so the chance of some individual developing resistance is higher and 2) People don't wash their hands long enough and well enough for the antibacterial soap to kill all the germs. Honestly, how many people you know that wash their hands at least for 1 minute and srub between their fingers and up to their forearms?
And then when antibacterial soap is needed, such as in a hospital dealing with a patient who's immune system is busted it would be desirable for the antibacterial soap to be very effective as not to expose the patient to that bacteria.
To most people such claims as "the virus is developing the ability to resist some drug" might sound as if the microorganisms can think and somehow train themselves to defend against drugs.
But what that means is that the virus undergoes mutations at a certain rate. And eventually one of those mutations might lead to resistance to the drug. All the other variants will be destroyed but that very small population which has the resistance will spread very rapidly.
That is why some say that not taking the full dose of antibiotics and using antibiotic soaps in homes can lead to the breeding of super-bacteria. A problem bigger than bird flu at this moment is antibiotic resistant staph bacteria (methicillin-resistant S. aureus aka MRSA). When you hear about people getting sicker just by being in the hospital - they probably caught MRSA. The deadliness and the number of cases from such infections have gone up even though it would make sense for them to go down with all the advances in medicine and hygiene. Some speculate that soon there will be another strain of staph bacteria (VRSA) vancomycin resistant S. aureus which would pretty much be resistant to all the known anti-biotics. All that has to happen is for microorganisms to mutate and spread faster than it takes for us to find new antibiotics.
Learn java. They are both object-oriented but with java it is easier to get started, there is more information and literature on Java and there are more examples and people who you can ask. Java will run well on any platform (no need to install Windows just for.NET or no need to install Mono or wait until Mono might catch up with the newest version of.NET.
Grab the JDK 1.5, grab a book or a tutorial, and install a good IDE (Netbeans, Eclipse) or just use a plain old text editor to keep it simple and you are set.
I know, but it doesn't come with Python by default and I wish it did. I want to tell a client to "just install Python" as opposed to "just install Python and version 2.6 of wxPython".
The post above was talking about Python coming with "the batteries included". Then I replied that one of the "batteries" is actually a "steam engine" - Tk/Tkinter. I wish Guido+python-dev would decide to use wxWindows, but they probably have their reasons that us mortals cannot understand;)
Re:Python vs. compiled Java and C
on
Guido Goes Google
·
· Score: 2, Insightful
I believe that "Premature optimization is the root of all evil." But most of all I believe in "the right tool for the right job".
Usually the first saying is applied when someone has decided on the language, say C, and on their hardware. But when programming, instead of thinking of the algorithm and how to correctly write it, they think about how many registers can they use at the same time first and start with inlining of assembly code. I think though, that the same principle can be applied to the choice of language and hardware. In this the mistake would be to choose a lower level language just because of speed, even though the problem can be solved _much_ faster (in terms of developing time) in a higher level language.
For example I had to write a little project to process text (about 2 million articles including some pdf files) then to do some scientific computation and finally design a reasonable GUI. The languages I new were C/C++/Java. I started with Java but I realized the project was just going to slow. I new what I wanted to get done, it is just it took a long time to write in Java. I looked around and tried a number languages Ruby, Perl, Python, even Matlab, tried to move some of the processing into MySQL as stored procedures. The best choice was to switch to Python. I didn't know Python at all but within a week I was fairly proficient with it. It is just as good of a OO language as Java and comes with a good set of supporting libraries. My project took about a couple of months and I wrote it in about 10K lines. I am estimating but I think I would have taken me twice as long to write it in Java and it would have been at least 15K or more lines long. The added problem with more code for the same problem is support and quality control. I can browse much faster through a 10k lines than through 20K lines to look for a bug or to some refactoring.
But all this doesn't mean that everyone, including you, should switch to Python immediately and drop C and C++. I acknowledge that Python is slower on average that other compiled (even JIT compiled) languages and some domains of application require speed most of all. As it turns out though, it is usually a vrey small part of code that needs to be sped up. In other words it is the good 'ol 80/20 rule (80% of the time the program is in the 20% of the code). What this means is that you could write the project faster in a higher level language like Python then choose certain critical part of your code (this requires having a head on the shoulders to know which parts are those and to know a little about algorithmic complexity O(n) stuff...) and optimize just that. This can be done beautifully in Python. As of now there at least 2 methods to do it: you can inline C code in Python using weave (hosted on the SciPy project page) or you can use a language extension to manipulate C data in Python using Pyrex. (I know there are other, but these are the most stable ones I think). I will use weave to inline some C code to speed up some of the computation. The result actually might be _faster!_ than Java.
In the end it all comes down to cost analysis, because you have a trade-off: for example, you can pay your developers twice as much and expect to pay twice for support and patches and have them develop in Java or you can achieve the same speed by allowing them to use Python (if they are smart enough they'll know how to optimize the critical regions after the program is runnig correctly) and you can buy extra(newer) hardware to compensate for the constant speed difference. If they can both optimize their Python code with C (in the end!) and buy faster hardware you might get a double benefit. But, of course, it is up to you (the company) to decide.
Python is just as fast as c or java when it comes to io-intensive applications:
That is because I/O is the bottleneck. In other words a program in C can execute its code in 1 ms but then have to wait 10 ms for next input you might as well write the code in Perl or Python and have it execute 9 ms and have it just wait 1 ms until next input. The programs, as far as the benchmark time is concerned will probably run at the same speed. In other words, I/O instensive applictions should be used mostly to test IO and not the programming languages.
Speaking of I/O performance. Is it possible I wonder to have an IO intensive Python script run faster than C? Actually it is possible. Python could have better caching mechanisms for example. Or _what if_ the interpreter could scan ahead and guess when/what the next IO will be and try to pre-fetch the data or prepare the output channel (sometimes speculatively) until the line is actually interpreted. So if I have a loop that fetches a line at each iteration.
for line in filehandle: processLine(line)
There might be some intelligent mechanism to anticipate and guess a repeated line fetch from the file and so 100 lines would be pre-fetched by the interpreter. The C compiler might not have this ability so it might seem that in certain cases IO in Python could be faster.
Sorry for the mistake, you are right, python is byte-compiled. I got JIT compiled to native confused with just bytecode compiled in my head (I know about Psyco but it is not used by default).
I really just guessed on the numbers of how much faster C or java bytecode would be than Python.
The reason I mentioned the constant multiple factor was that the other alternative could have been for example an exponential difference (I'll give an example below). Let's say printing to console in python takes 5 times longer than printing to console in C. Now the hardware is upgraded and the speed of the CPU is about twice as fast as before. But printing in python still takes 5 times longer than printing in C. After a couple of such upgrades (which should come about once in 18 months according to the good 'ol Moore's law) Python will run as fast as C ran just 2 or so upgrades ago. That is good news for Python. The stuff that a while ago just had to written in C for purely CPU performance (not taking into account IO bottlenecks) can now be written in Python. This is true in genaral for computers we use today. Because they are all basically Turing machines.
But it turns out that in the case of a quantum computer (if there was one functioning) and the problem of factoring, this would not have happened. If, I had to factor a certain number. A classic machine (a turing machine) could do it X times slower than a quantum computer. Now the speed of the PC is doubled (say the # of bits in the registers, bus, and memoory is doubled) and the speed of the quantum computer is also doubled (double the # of quantum bits) then the quantum computer might be X*X=X^2 faster. On the next similar hardware upgrade it would be X*X*X=X^3 faster and so on. That would be "bad news" if there was any intention of eventually solving some factoring problem in the future with a classical machine that the quantum one can solve today - it means it will _never_ be able to catch up. It would have been as if certain problems that need C today would never be able to be solved just as fast in Python not matter how much faster the classical computer will be upgraded.
Printing a string to the console (aka the "hello world" problem) is something done often in programming. Most programs out there have to send output to an output device.
I, like many other people out there, want to see this "hello world" problem solved in any new language that I am learning for illustration purposes and because it gives a feel about the language, for example: what are the delimiters?, are any variables declared?, are there any object being instantiated?, is there a need to import any libraries for I/O? and of course, how many lines would be needed to do a basic task of outputting a string to a console? Most computer books that introduce a new language, if you have seen any, will include these type of programs and for good reasons.
In this particular case the hello world program doesn't show that Python is the absolute best language out there, but it does show that even to print a string to the console java needs 3 or 4 more time the ammount of code that Python does. The implication is that _in general!_ to solve the same problem in java one would need to type more lines of code. I could have shown you a 2000+ lines little project I wrote and the n an almost similar solution written in Java that took about 3000+ lines, but don't you think that would have been a little too much?
But since you brought your example in an attempt to show how stupid the "print "text" example is, let's see what one would learn just from it anyway. So we have:
print "text"
then "python file.py"
as opposed to:
10 PRINT "TEXT"
then GWBASIC FILE.BAS
Just from that I can guess that both are probably interpreted, both are high level (no need to allocate memory, or set up interrupts and load anything into registers), both languages are probably not strictly object oriented because there are no objects being created, both have built-in statements for console output, strings are enclosed by double quotes. also no visible end-of-line or end-of-statement characters are there, so whitespace might be used as a delimiter. Now what is different about them: Because one has a line # number I assume that it probably relies heavily on the GOTO statement and the other one doesn't. Now that is quite a bit learned just from 2 "stupid" "hello world" examples...
Re:Semi-Off-Topic Python vs. Perl discussion
on
Guido Goes Google
·
· Score: 2, Interesting
I switched from Java to Python. Today's hardware probably runs Python just as fast as the hardware 5 years ago ran Java. My reasons why I like like Python in decreasing order (some of them you have mentioned):
1. OO - with classes, inheritance and other OO goodies 2. scripted - no need to compile, easy to debug 3. powerful native data types - dictionaries (hashes), lists and sets(new in 2.xx) 4. simple syntax (indentation => code blocks) - a blessing in disguise, I can read anybody's Python program - the format is always the same 5. rich library and extensions - networking, GUI(wxWindows, Qt, Gtk, Tk...),math (SciPy, Numeric), regex, file and string manipulation etc. 6. can inline C code with weave, can manipulate C data in Python with Pyrex, can use JIT compile with Psyco 7. can use threads 8. is a completely free and open source unlike Sun's Java 9. has some functional/declarative constructs (lambda, map, reduce and apply) 10. i like the name Guido - it sounds cool (ok, ok I don't really know how to say it but I needed the 10th reason...)
Python vs. compiled Java and C
on
Guido Goes Google
·
· Score: 4, Interesting
The ability to inline C code in Python rocks.
I like weave but I am waiting to see what will come out of the PyPy project - the author is the one who wrote Psyco (the JIT compiler for Python).
Then there is Pyrex where Python can manipulate C data with language extensions, as opposed to weave where C code is inlined into the python code and Python data is manipulated in C.
It is true that as a scripting language Python is slower than (byte)compiled languages. But it is slower by a constant factor. In other words people would say "Your Python solution is 4x slower than my Java solution!". What this means though is that just by upgrading the hardware that Python runs on, one can reach the speed of execution of the compiled program. In other words Python on an Athlon 64 fx 57 might run faster than java bytecode on a 1Ghz Athlon, or might even run faster than C on a 100Mhz machine (I am just making these numbers up, maybe someone knows of some benchmarks?). The point is that application that required C 15-20 years ago can probably be re-written in Python now.
have a robust low-level language generating high-level language (HLL) code
What are some examples? I can easily think of a high level language generating a lower-level like gcc transforming your C++ into machine language, or java being compiled into bytecode instructions.
As for a HLL->HLL translation take a look at the PyPy project (which among other things is attempting to write a Python interpreter in Python). here
The goal is to generate low level description of the python code that interprets the python code (go figure!). That is what they claim at least...
You said that what was presented was not "code generation" but "application generation", but you never explained what the difference is. What should the above fragment look like to make it "code generation"?
Not all the batteries are the same. For example I like to think of Tk(Tkinter) not as a battery but more as steam engine - clunky, odd looking, slow and hard to use (I don't know why wxWindows hasn't replaced it yet) - besides that I really like Python.
Besides coffee, when people say Java they sometimes mean different things:
For some it is the byte compiled oo language, these people chose it because it looked like C++ but with no explicit pointers and a GC.
For some it is the platform, in other words they liked the large library of classes that came with it like swing/awt, math, networking.
Others liked it because of the technologies built on top of it or with it: J2EE (EJBs etc.), Applets, Servelets
Yet others liked it just because it is the most portable and functional language+platform they had ever seen. Here when a company makes a product written in C/C++ that they have to port to another platform, having a "compile once, run anywhere" language can cut costs of developing/maintenance dramatically
As far as the Java fanatics and cheerleader of the past years are concerned they probably moved on because that is what they do - they get excited about a new technology for a while, make a lot of noise then move to another new technology. Java is already used extensively in the industry and is not going away soon.
As far as Ruby and Python. I would have to say that Python is probably the best successor of Java. Why not Ruby? Because, as you say Ruby is a raw young language it doesn't have nearly the libraries that Java and Python has. The only context I ever hear of Ruby is when it is mention in the "Ruby on Rails" title. I know it is a more pure 'cleaner' object oriented language than PYthon, Python is just more practical, and more intuitive as far as I am concerned. I was planning to write my project in Java, but after a week of playing with Python, I was up and running and I believe I finished in 2/3 or 1/2 the time it would have taken me to write it in Java. Look at a simple code comparison in java and one in Python
:
In Java
public class J { public static void main(String[] args){ System.out.println("text"); } }
then you have to run: "javac J.java" then "java J"
Versus the Python version
print "text" then to run just: "python file.py"
Ruby is pretty close but I dismissed it as not having enough libraries and add-ons. With python for example, I can even run it on a Java platform with JPython and still access the rich java libraries. I can optimize parts of my code in C and either use the syntax extensions of Pyrex, or weave to inline that code into regular python code. There are good GUI bindings (Qt, wxWindows, Gtk and of course the default Tk, for those seeking to punish themselves), there is threading, networking, good regex support. My favorite is the native dictionary (hash table) data type. Just use something like "dic={}" to initialize an emtpy dictionary then "dic[key]=value" to use it, and that is it - no need for new no need to import anything.
For me it is more a matter of who controls the money. If you are the artist I will want to pay _you_ if I like _your_ work. And then however _you_ or _your_ manager decide to spend it, it doesn't matter.
What I see happening now is that major media production companies will sign some contract with an artist that they should release so much music of so an so kind and get only this much of total profit - many have no choice and sign it. I understand that one needs a large capital to even kickstart a major producion, but if an artist feels that want to produce so much they are probably already well-known and have enough funds.
I just want to point out that the magazine analogy and music doesn't work. Digital products are not the same as tangible material products.
A photocopy of a magazine would take a long time and the quality would be pretty bad - in other words I would have to part with my magazine for a long time to let my friend photocopy it and the quality of graphics he would get would be far worse. Also photocopying could cost quite a bit. It would be in our both interests for him to go buy the magazine.
With a CD (data or music) I can make an identical copy (as far as quality is concerned) in under 3 minutes. So 1) I don't part with my copy of the product. 2) My friend gets an identical product 3) The cost of blank CDs is really low (I buy them in packs of 100). In this case it would not make any sense to send my friend to the store and have him pay $20 for music if I can make a copy for him.
As far as SD/MMC go, some of them suffer from dead bits - so over time the bits would go 'bad'. I am not sure how it compares with a CD but I know it happens. In case of a handheld PC I had (Zaurus) I used to mount the compact flash card with the noatime option so every time the files where accessed (even just for reading) a timestamp wouldn't be written back to the media because there is a limited time a single bit could be re-written before it goes 'bad'.
Actually I would really like if artists had an option of donating money. I would go and just donate to the artist.
And I don't think pirating is right, but I don't like the scams run by the recording studios ($20 for disk is a scam if you ask me). So I have not bought a single CD for about 3 years now. I got some music off iTunes, in rest I downloaded free remixes of songs of some artists I like and then listened to the radio and to my old CDs.
Slashdot Mirror
Well, I noticed my CDMA phone doens't do that and I also noticed the interference is "loudest" when I bring the phone close to the computer case. But you are probably right, I really don't know that much about electronics...
How about someone that graduated from a 4 year (1 year co-op=5 year) US university with the Computer Science degree and a 3.94 GPA? Oh, and I am a foreigner -- I just happen to have a green card?
Then switched to Verizon and got a free basic LG VX3200 and now I have a _phone_ that works as a phone. The bottom line is I am not buying a phone only - I am buying a phone with a service (network that comes with it). Verizon so far is the best in my area, everyone I know that has either Cingular or T-Mobile has crappy reception.
Also, what is up with GSM phones that they interfere with the electronics in my computer - when the phone was on my desk it would periodically induce this clicking noise in my speakers...
The more people they scare the better. This is terrorism at its best. It is not that they will be able to sue everyone who ever accessed a P2P network, but they sure can terrorize people like with just a couple of "examples" like this and then also do batch litigation a 1000 or more people at a time here and there -- just enough to have everyone's "friend of friend" go through this. When stuff happens to "people you know" then one really starts thinking that it might happen him/her.
I still consider the best strategy if one gets caught is to infect your own machine with viruses/worms/spyware/trojans - anything that will provide a possible explanation of how the files ended up on there
If you want to talk about viruses only, reply to a different comment and use the word "evolution" as many times as you want in your post. Nobody is going to stop you (aside from almighty moderators, of course;)
For unclear reasons you are desperate to have the word "evolution" some place in the comment so here you go: EVOLUTION, better now?
Quoting Wikipedia is the same as quoting yourself, since you could have technically gone and edited the entry to support your point of view.
You are right, Ruby is cleaner and more consistent. But Python works for me because it has more 3rd party mechanisms to inline C code and thus I can optimize say 3 nested for loops in a certain function. Ruby has been popular in Japan and Asia for quite a while but I think there is more documentaion and examples of code for Python.
I don't think it is considered evolution in the classic sense. Just because you really, really want to use the word "to stick it" to those crazy people from Kansas doesn't mean that you should. What happens is that bacteria has a specific set of genes that will mutate faster and those genes will protect it from certain molecules. So we have seen staph mutate to protect itself against different antibiotics and that can happen very quickly but staph will probably not turn into a fungus or an insect just as quickly. It seems the bacteria have evolved this fast mutating mechanism of quick adaptability to defend itself agianst toxins but it is limited to that. But the ability to have this kind of defense mechanism would be evolution in the classic sense.
and then from the same article
The point is that antibacterial soap should have its uses and is recommended to be used in the hospitals. But when it is used at home and everywhere eventually the bacteria will develop resistance., because 1) A lot more bacteria is exposed to it so the chance of some individual developing resistance is higher and 2) People don't wash their hands long enough and well enough for the antibacterial soap to kill all the germs. Honestly, how many people you know that wash their hands at least for 1 minute and srub between their fingers and up to their forearms?
And then when antibacterial soap is needed, such as in a hospital dealing with a patient who's immune system is busted it would be desirable for the antibacterial soap to be very effective as not to expose the patient to that bacteria.
But what that means is that the virus undergoes mutations at a certain rate. And eventually one of those mutations might lead to resistance to the drug. All the other variants will be destroyed but that very small population which has the resistance will spread very rapidly.
That is why some say that not taking the full dose of antibiotics and using antibiotic soaps in homes can lead to the breeding of super-bacteria. A problem bigger than bird flu at this moment is antibiotic resistant staph bacteria (methicillin-resistant S. aureus aka MRSA). When you hear about people getting sicker just by being in the hospital - they probably caught MRSA. The deadliness and the number of cases from such infections have gone up even though it would make sense for them to go down with all the advances in medicine and hygiene. Some speculate that soon there will be another strain of staph bacteria (VRSA) vancomycin resistant S. aureus which would pretty much be resistant to all the known anti-biotics. All that has to happen is for microorganisms to mutate and spread faster than it takes for us to find new antibiotics.
Grab the JDK 1.5, grab a book or a tutorial, and install a good IDE (Netbeans, Eclipse) or just use a plain old text editor to keep it simple and you are set.
The post above was talking about Python coming with "the batteries included". Then I replied that one of the "batteries" is actually a "steam engine" - Tk/Tkinter. I wish Guido+python-dev would decide to use wxWindows, but they probably have their reasons that us mortals cannot understand ;)
Usually the first saying is applied when someone has decided on the language, say C, and on their hardware. But when programming, instead of thinking of the algorithm and how to correctly write it, they think about how many registers can they use at the same time first and start with inlining of assembly code. I think though, that the same principle can be applied to the choice of language and hardware. In this the mistake would be to choose a lower level language just because of speed, even though the problem can be solved _much_ faster (in terms of developing time) in a higher level language.
For example I had to write a little project to process text (about 2 million articles including some pdf files) then to do some scientific computation and finally design a reasonable GUI. The languages I new were C/C++/Java. I started with Java but I realized the project was just going to slow. I new what I wanted to get done, it is just it took a long time to write in Java. I looked around and tried a number languages Ruby, Perl, Python, even Matlab, tried to move some of the processing into MySQL as stored procedures. The best choice was to switch to Python. I didn't know Python at all but within a week I was fairly proficient with it. It is just as good of a OO language as Java and comes with a good set of supporting libraries. My project took about a couple of months and I wrote it in about 10K lines. I am estimating but I think I would have taken me twice as long to write it in Java and it would have been at least 15K or more lines long. The added problem with more code for the same problem is support and quality control. I can browse much faster through a 10k lines than through 20K lines to look for a bug or to some refactoring.
But all this doesn't mean that everyone, including you, should switch to Python immediately and drop C and C++. I acknowledge that Python is slower on average that other compiled (even JIT compiled) languages and some domains of application require speed most of all. As it turns out though, it is usually a vrey small part of code that needs to be sped up. In other words it is the good 'ol 80/20 rule (80% of the time the program is in the 20% of the code). What this means is that you could write the project faster in a higher level language like Python then choose certain critical part of your code (this requires having a head on the shoulders to know which parts are those and to know a little about algorithmic complexity O(n) stuff...) and optimize just that. This can be done beautifully in Python. As of now there at least 2 methods to do it: you can inline C code in Python using weave (hosted on the SciPy project page) or you can use a language extension to manipulate C data in Python using Pyrex. (I know there are other, but these are the most stable ones I think). I will use weave to inline some C code to speed up some of the computation. The result actually might be _faster!_ than Java.
In the end it all comes down to cost analysis, because you have a trade-off: for example, you can pay your developers twice as much and expect to pay twice for support and patches and have them develop in Java or you can achieve the same speed by allowing them to use Python (if they are smart enough they'll know how to optimize the critical regions after the program is runnig correctly) and you can buy extra(newer) hardware to compensate for the constant speed difference. If they can both optimize their Python code with C (in the end!) and buy faster hardware you might get a double benefit. But, of course, it is up to you (the company) to decide.
That is because I/O is the bottleneck. In other words a program in C can execute its code in 1 ms but then have to wait 10 ms for next input you might as well write the code in Perl or Python and have it execute 9 ms and have it just wait 1 ms until next input. The programs, as far as the benchmark time is concerned will probably run at the same speed. In other words, I/O instensive applictions should be used mostly to test IO and not the programming languages.
Speaking of I/O performance. Is it possible I wonder to have an IO intensive Python script run faster than C? Actually it is possible. Python could have better caching mechanisms for example. Or _what if_ the interpreter could scan ahead and guess when/what the next IO will be and try to pre-fetch the data or prepare the output channel (sometimes speculatively) until the line is actually interpreted. So if I have a loop that fetches a line at each iteration.
There might be some intelligent mechanism to anticipate and guess a repeated line fetch from the file and so 100 lines would be pre-fetched by the interpreter. The C compiler might not have this ability so it might seem that in certain cases IO in Python could be faster.Or say add an "int main()" ... plus may be some curly braces here and there..
But how about a litte compilation and linking? perhaps even a Makefile...
I really just guessed on the numbers of how much faster C or java bytecode would be than Python.
The reason I mentioned the constant multiple factor was that the other alternative could have been for example an exponential difference (I'll give an example below). Let's say printing to console in python takes 5 times longer than printing to console in C. Now the hardware is upgraded and the speed of the CPU is about twice as fast as before. But printing in python still takes 5 times longer than printing in C. After a couple of such upgrades (which should come about once in 18 months according to the good 'ol Moore's law) Python will run as fast as C ran just 2 or so upgrades ago. That is good news for Python. The stuff that a while ago just had to written in C for purely CPU performance (not taking into account IO bottlenecks) can now be written in Python. This is true in genaral for computers we use today. Because they are all basically Turing machines.
But it turns out that in the case of a quantum computer (if there was one functioning) and the problem of factoring, this would not have happened. If, I had to factor a certain number. A classic machine (a turing machine) could do it X times slower than a quantum computer. Now the speed of the PC is doubled (say the # of bits in the registers, bus, and memoory is doubled) and the speed of the quantum computer is also doubled (double the # of quantum bits) then the quantum computer might be X*X=X^2 faster. On the next similar hardware upgrade it would be X*X*X=X^3 faster and so on. That would be "bad news" if there was any intention of eventually solving some factoring problem in the future with a classical machine that the quantum one can solve today - it means it will _never_ be able to catch up. It would have been as if certain problems that need C today would never be able to be solved just as fast in Python not matter how much faster the classical computer will be upgraded.
I, like many other people out there, want to see this "hello world" problem solved in any new language that I am learning for illustration purposes and because it gives a feel about the language, for example: what are the delimiters?, are any variables declared?, are there any object being instantiated?, is there a need to import any libraries for I/O? and of course, how many lines would be needed to do a basic task of outputting a string to a console? Most computer books that introduce a new language, if you have seen any, will include these type of programs and for good reasons.
In this particular case the hello world program doesn't show that Python is the absolute best language out there, but it does show that even to print a string to the console java needs 3 or 4 more time the ammount of code that Python does. The implication is that _in general!_ to solve the same problem in java one would need to type more lines of code. I could have shown you a 2000+ lines little project I wrote and the n an almost similar solution written in Java that took about 3000+ lines, but don't you think that would have been a little too much?
But since you brought your example in an attempt to show how stupid the "print "text" example is, let's see what one would learn just from it anyway. So we have:
then "python file.py"as opposed to:
then GWBASIC FILE.BAS
Just from that I can guess that both are probably interpreted, both are high level (no need to allocate memory, or set up interrupts and load anything into registers), both languages are probably not strictly object oriented because there are no objects being created, both have built-in statements for console output, strings are enclosed by double quotes. also no visible end-of-line or end-of-statement characters are there, so whitespace might be used as a delimiter. Now what is different about them: Because one has a line # number I assume that it probably relies heavily on the GOTO statement and the other one doesn't. Now that is quite a bit learned just from 2 "stupid" "hello world" examples...
I like weave but I am waiting to see what will come out of the PyPy project - the author is the one who wrote Psyco (the JIT compiler for Python).
Then there is Pyrex where Python can manipulate C data with language extensions, as opposed to weave where C code is inlined into the python code and Python data is manipulated in C.
It is true that as a scripting language Python is slower than (byte)compiled languages. But it is slower by a constant factor. In other words people would say "Your Python solution is 4x slower than my Java solution!". What this means though is that just by upgrading the hardware that Python runs on, one can reach the speed of execution of the compiled program. In other words Python on an Athlon 64 fx 57 might run faster than java bytecode on a 1Ghz Athlon, or might even run faster than C on a 100Mhz machine (I am just making these numbers up, maybe someone knows of some benchmarks?). The point is that application that required C 15-20 years ago can probably be re-written in Python now.
What are some examples? I can easily think of a high level language generating a lower-level like gcc transforming your C++ into machine language, or java being compiled into bytecode instructions.
As for a HLL->HLL translation take a look at the PyPy project (which among other things is attempting to write a Python interpreter in Python). here The goal is to generate low level description of the python code that interprets the python code (go figure!). That is what they claim at least...
You said that what was presented was not "code generation" but "application generation", but you never explained what the difference is. What should the above fragment look like to make it "code generation"?
Not all the batteries are the same. For example I like to think of Tk(Tkinter) not as a battery but more as steam engine - clunky, odd looking, slow and hard to use (I don't know why wxWindows hasn't replaced it yet) - besides that I really like Python.
For some it is the byte compiled oo language, these people chose it because it looked like C++ but with no explicit pointers and a GC.
For some it is the platform, in other words they liked the large library of classes that came with it like swing/awt, math, networking.
Others liked it because of the technologies built on top of it or with it: J2EE (EJBs etc.), Applets, Servelets
Yet others liked it just because it is the most portable and functional language+platform they had ever seen. Here when a company makes a product written in C/C++ that they have to port to another platform, having a "compile once, run anywhere" language can cut costs of developing/maintenance dramatically
As far as the Java fanatics and cheerleader of the past years are concerned they probably moved on because that is what they do - they get excited about a new technology for a while, make a lot of noise then move to another new technology. Java is already used extensively in the industry and is not going away soon.
As far as Ruby and Python. I would have to say that Python is probably the best successor of Java. Why not Ruby? Because, as you say Ruby is a raw young language it doesn't have nearly the libraries that Java and Python has. The only context I ever hear of Ruby is when it is mention in the "Ruby on Rails" title. I know it is a more pure 'cleaner' object oriented language than PYthon, Python is just more practical, and more intuitive as far as I am concerned. I was planning to write my project in Java, but after a week of playing with Python, I was up and running and I believe I finished in 2/3 or 1/2 the time it would have taken me to write it in Java. Look at a simple code comparison in java and one in Python
: In Java
Versus the Python versionRuby is pretty close but I dismissed it as not having enough libraries and add-ons. With python for example, I can even run it on a Java platform with JPython and still access the rich java libraries. I can optimize parts of my code in C and either use the syntax extensions of Pyrex, or weave to inline that code into regular python code. There are good GUI bindings (Qt, wxWindows, Gtk and of course the default Tk, for those seeking to punish themselves), there is threading, networking, good regex support. My favorite is the native dictionary (hash table) data type. Just use something like "dic={}" to initialize an emtpy dictionary then "dic[key]=value" to use it, and that is it - no need for new no need to import anything.What I see happening now is that major media production companies will sign some contract with an artist that they should release so much music of so an so kind and get only this much of total profit - many have no choice and sign it. I understand that one needs a large capital to even kickstart a major producion, but if an artist feels that want to produce so much they are probably already well-known and have enough funds.
A photocopy of a magazine would take a long time and the quality would be pretty bad - in other words I would have to part with my magazine for a long time to let my friend photocopy it and the quality of graphics he would get would be far worse. Also photocopying could cost quite a bit. It would be in our both interests for him to go buy the magazine.
With a CD (data or music) I can make an identical copy (as far as quality is concerned) in under 3 minutes. So 1) I don't part with my copy of the product. 2) My friend gets an identical product 3) The cost of blank CDs is really low (I buy them in packs of 100). In this case it would not make any sense to send my friend to the store and have him pay $20 for music if I can make a copy for him.
As far as SD/MMC go, some of them suffer from dead bits - so over time the bits would go 'bad'. I am not sure how it compares with a CD but I know it happens. In case of a handheld PC I had (Zaurus) I used to mount the compact flash card with the noatime option so every time the files where accessed (even just for reading) a timestamp wouldn't be written back to the media because there is a limited time a single bit could be re-written before it goes 'bad'.
And I don't think pirating is right, but I don't like the scams run by the recording studios ($20 for disk is a scam if you ask me). So I have not bought a single CD for about 3 years now. I got some music off iTunes, in rest I downloaded free remixes of songs of some artists I like and then listened to the radio and to my old CDs.