Wikipedia:Reference desk/Archives/Computing/2008 June 3

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Computing desk
< June 2 << May | June | Jul >> June 4 >
Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

June 3[edit]

Internet Problems Following Failed Ubuntu Installation[edit]

Hello, I'm having problems accessing the internet which are pretty strange to me. Here's the situation:

I attempted to install an 8.04 Ubuntu release using a CD and got a short ways into the installation before I got a few error messages and a lot of lagging. I already knew my system only had a little under 256 mb of memory when more was generally required, and I was going to install Xubuntu instead so I didn't give it much thought. I unfortunately did a hard reset (which I know is always a stupid idea, but I couldn't get the system to pause or close by combinations of escape and other keys) and tried to start up Windows. This was generally fine except for Windows asking to check the disk, which I allowed. This also went fine, but there was some message following the 3rd step that I couldn't really follow because it closed quickly.

When I signed back on to my account, all my files were in order but the internet was, and still is, acting up. I'm pretty sure I can access sites without a problem- given enough time. As I type this the icons for different buttons are still loading, and internet browsing in general seems to be rather slow to work. At first it was, strangely, only Wikipedia that didn't load. In fact, quite a few sites are still rather quick to load, but others just hang for a long time almost without rhyme or reason.

My real question is, what could a hard reset during the attempt at installation (I should note that I chose to install Ubuntu without changing the existing configuration) mess up that would leave internet access, files, and the like, while slowing down only certain sites at certain times? —Preceding unsigned comment added by (talk) 00:53, 3 June 2008 (UTC)

Request for more informationHow are you connected to the Internet? Dial-up? Ethernet Cable? DSL? Wi-Fi? Are you a poweruser (admin) on Windows? Are you sure it is the Internet connection, and not the web browser, that is acting strangely? AFAIK, if you are not a poweruser and some of your browser files accidentally got corrupted during a hard boot, Windows XP SP2 would not allow the browser (most probably, Mozilla Firefox) to set it correctly. Of course, I could be off a tangent with the question, too. Kushal (talk) 01:09, 3 June 2008 (UTC)

Sorry that I wasn't specific:

1) I have a cable connection and I'm using a router, but it hasn't been changed at all by the setup, and a computer in another room actually seems to be fine.

2) I'm pretty sure it isn't the web browser (Firefox) because I tried accessing sites on Internet Explorer, too.

3) Someone in my house said it might be a system settings problem too, which I'm starting to think it is. I tried to quickly fix the internet connection by doing a Windows repair but nothing changed...

Is there any way I would be able to figure out the source of the problem and fix it? On another note, when I tried to access this page to edit it I opened another tab, assuming that it might load faster. When the page in fact loaded, both of the windows reached the website at the same time. Could this mean that the internet is working in bursts or something?

Also, I am connected to an instant messaging program with absolutely no problems. —Preceding unsigned comment added by (talk) 01:20, 3 June 2008 (UTC)

As far as I know, Internet access always works in bursts. Kushal (talk) 03:59, 3 June 2008 (UTC) Are you a poweruser (or administrator) on your computer? Kushal (talk) 04:01, 3 June 2008 (UTC)

IPhone or IPod touch?[edit]

I'm not sure if this is the right section but anyway... I'm thinking on getting either and Iphone or an Ipod touch. Which one do you think I should get?

The Iphone has more app. (like the phone, camera ect.) but, the Ipod touch is faster, cheaper and comes with more memory. —Preceding unsigned comment added by (talk) 02:09, 3 June 2008 (UTC)

If you want a phone, consider the iPhone, otherwise the iPod Touch. Or just get a Razr or some other cheap/decent phone and a Touch. Or whatever your heart desires. Though some might criticize the Touch's low capacity, I know one guy who loves his. It fits more than a day's worth of music anyways, you just need to swap it out every once in awhile (and if you had a particularly huge collection you would be doing this anyways) -- Consumed Crustacean (talk) 03:02, 3 June 2008 (UTC)
I wonder if a 3G (HSDPA or otherwise) iPhone with 16 GB or more storage capacity is coming this July. I agree with Consumed Crustacean. If you do not want a phone, save the money that you would spend on a phone. If you need a phone and not an iPod, get yourself a phone. Before you decide the iPod Touch over the iPhone, make sure you understand that there are certain things that current generation iPod Touch devices can never do (and that do not seem to matter yet, vibration may be one of them). There are many rumors, and according to one, you might get an iPhone for as little as USD 200.00 as an initial payment.
Oh, by the way, have you heard of Android (mobile device platform)? maybe that could be your next phone in two years time after you get your iPhone 2 this fall? Kushal (talk) 03:57, 3 June 2008 (UTC)

What to learn next in programing[edit]

I've been on and off with programming - a little PHP and HTML, and some Perl from a while back - like loops, variables, functions. I worked a lot with Visual Basic and access, and I know a good amount of how do use: databases, OO programming(with VB- easy stuff), using IDEs, and well, basic stuff in VB). A couple of weeks back, I picked up PHP again, and I've learned quite a bit with PHP/MySQL: arrays, function, databases, post/get, cookies, date, and file manipulation. I was considering on taking up a project with PHP and MySQL, but I've been having second thoughts about it, because It's a really time consuming one, requiring AJAX (javascript + php - really a pain to use), which might teach me a lot about AJAX, but I don't think it would teach me things any deeper than I know already.

So, I'm at the point in programming where I feel that I could do a lot of things in a programming language (using tons of reference), but I'm not sure exactly what to learn/pursue next, however, if it makes sense.

So, I have a couple of questions, if you don't mind (sorry to make it so long!):

1) Would I still be considered "new" to programming based on the stuff I learned?
2) Should I drop my PHP project, and learn other stuff?
3) Is it bad to be skipping from programming language to programming language?
4) What should I learn next (Java/C++/Python, etc.)?
—Preceding unsigned comment added by (talk) 04:46, 3 June 2008 (UTC)

It is more a software design approach than straight programming, but I would suggest learning UML and Design Patterns if you haven't (doing both together would probably be a good combination, but start with the basics of UML to help make sense of Design Patterns). Using them to design your Javascript -> PHP -> SQL project would be a great help and a perfect learning experience in my opinion. And as far as the actual programming goes, I wouldn't say jumping between languages is a bad thing as long as it doesn't confuse you too badly, so that'll have to be your call. --Prestidigitator (talk) 05:52, 3 June 2008 (UTC)

If you're interested in doing more with web programming, Django is a beautiful system based on Python, which itself is a nice language to learn. Django is object-oriented and deeply integrated with SQL, and it has some modules that do AJAX for you, so this could solve your PHP woes. Of course, there's also the ever-popular Ruby on Rails for that purpose -- it has more AJAX magic, but I've been burned by its inconsistency.

It's also never a bad thing to skip from one programming language to another -- it will eventually let you spend more of your brainpower on general, useful programming patterns and less on the mundane specifics of each language. On that note, if you want to become really good at programming, you could try a language that's very different than the ones you've learned, to expand your repertoire. A functional language such as LISP or Scheme would do well for that purpose. (I was almost going to suggest Haskell, but I realized that it is frankly a difficult, mind-bending language, and would be prone to making you give up and return to the comfort of PHP or VB.)

Finally, here are some good programming books that you can read online for free:

rspeer / ɹəədsɹ 09:47, 3 June 2008 (UTC)\

Thanks for your replies! So, I'm wondering which language to move onto. C++, Java, or Python. I'm not really sure. I liked the "How to think like a computer programmer", but i was thinking about doing Java/C++ also. Would you reccomend OO/GUI programming for me? I'll be sure to check out some of the other links everyone has posted. Thanks again.
At the risk of being a bit too opinionated and starting a Wiki flame war, I'd say there's currently no reason to learn C++ unless you have to work with other people's C++ code. It's a language that tries to do everything with nothing, and ends up just doing everything in confusingly idiosyncratic ways. For example, C++ was designed to add the object-oriented paradigm to C, but we have a much better understanding of object-oriented design now than when C++ was created, so its templates and pointers look and feel positively hackish now. C++ will bite you in unexpected ways, because it has so many unexpected things in its specifications, which themselves fill a 1500-page book.
Java is fine. It's a very practical language, especially for the business world. It's the most portable of any of the languages you've mentioned, so if you want to write things for other people to run on all kinds of computer systems, it would probably be your best choice. But I have also found that it makes some aspects of programming quite dull. You may find yourself typing the same lines of boilerplate over and over and generally feeling like a corporate drone.
So, if you're programming on your own terms, I would definitely recommend Python out of those three. Python can be very intuitive to learn and fun to use, and once you've learned it, you can also learn Django and start making web applications with it. rspeer / ɹəədsɹ 22:20, 3 June 2008 (UTC)
Just to step in here, I agree with the above, but I'd put Java above Python. Depends a lot on what you want to do, but Java sticks to the c-type syntax that PHP uses (which makes the syntax easier, at least), and will force you to learn OO. OO is one of those big steps you really need to pass through in order to get good skills at software design. Python uses it, but Java insists. :) We teach Java, and my argument is that we don't teach it because Java itself is a good language (it is ok, but there are better languages) but because of what it forces you to learn and because of its application. Get some better OO design skills, and your Python/PHP/.Net etc programming will leap ahead. :) But, as Rspeer mentioned, Java is a hoop you should jump through for professional programming, while Python is better for hackers and others who want to code on their own terms, and is a beautiful language in its own right (noting that Python is used professionally, but not to the level that Java is). I'd add that knowing some C or C++ will help career-wise, but depending on where you want to go it might not be essential. - Bilby (talk) 04:05, 4 June 2008 (UTC)
Ok. Thanks! Between Scheme and Lisp, what do you recommend I learn? Are there any good IDEs/interpreters worth noting for either of them? I now plan on learning Python(most probably), along with Lisp or Scheme (basically, any low level languages), to get some experience with it. —Preceding unsigned comment added by Legolas52 (talkcontribs) 23:16, 3 June 2008 (UTC)
Ultimately, you will probably want to turn your talents into a career. I suggest you take a look at some IT recruitment websites and see what programming language skills employers are looking for. Astronaut (talk) 01:00, 4 June 2008 (UTC)
Scheme is more modern than LISP, and DrScheme is a good interpreter for it. But neither Scheme or Lisp would typically be considered low level, as in running things on the bare hardware, since modern computer hardware is really designed for C-like languages. So if you want to squeeze raw speed out of your programs, you probably want to know some C eventually. (Using C and Python together can be a particularly powerful way to program.) I recommended Scheme because it's useful to see a different perspective on programming -- Scheme is based on a paradigm where the fundamental unit is the function rather than the command. rspeer / ɹəədsɹ 03:03, 4 June 2008 (UTC)
The only thing I had to use Scheme for was a course based on recursion. Other than that, I've almost forgotten about it. I'm not sure I'd recommend it myself though, because although it is very good for some things, it's also a very small slice of the whole, imho. --Wirbelwindヴィルヴェルヴィント (talk) 03:49, 4 June 2008 (UTC)
Go for the theory: The art of computer programming.GoingOnTracks (talk) 23:04, 6 June 2008 (UTC)

Python sockets - sending from a specific IP address[edit]

I'd like to send some data in Python from a specific IP address, but want to let Python automatically choose a port to send from (because I could have many of these running at once). This is what I have so far - this sends from the IP address fine:

from socket import *
s = socket( AF_INET, SOCK_STREAM )
s.bind(( 'x.x.x.2', 2020 ))               # Sending data from a specific IP address, using a specific high port
s.connect(( '', 80 ))
s.send( 'GET / HTTP/1.1\r\nHost:\r\n\r\n' )
while True:
	d = s.recv(100)
	if d:
		print d

But if I bind with an IP address only with s.bind(( 'x.x.x.2' )) it sends it from the default IP address (x.x.x.1), not the one I specify. Is there any way of automatically allocating an IP address but not a port a port but not an IP address? --h2g2bob (talk) 11:02, 3 June 2008 (UTC)

Just like in straight C sockets, binding to port zero means "choose a port for me": s.bind(( 'x.x.x.2', 0 )) --Sean 13:37, 3 June 2008 (UTC)
When I do that, it goes back to the default IP address too :( Is there any way to fix the IP address but have the port allocated automatically? --h2g2bob (talk) 14:53, 3 June 2008 (UTC)
It works for me on my Linux machine, using the following server and client scripts to verify what IP the client is coming from:
# Server:
from socket import *
s = socket( AF_INET, SOCK_STREAM )
s.bind(( '', 9001 ))
print "Waiting for clients ..."
while True:
        cli = s.accept()
        print "Got client:",
        print cli

# Client:
from socket import *

server_ip  = ''

client_ip1 = ''
sock1 = socket( AF_INET, SOCK_STREAM )
sock1.bind(( client_ip1, 0 ))
sock1.connect(( server_ip, 9001 ))

client_ip2 = ''
sock2 = socket( AF_INET, SOCK_STREAM )
sock2.bind(( client_ip2, 0 ))
sock2.connect(( server_ip, 9001 ))

# Running gives:
# Waiting for clients ...
# Got client: (<socket._socketobject object at 0x4037a2d4>, ('', 60709))
# Got client: (<socket._socketobject object at 0x4037ee8c>, ('', 60710))
If it doesn't work for you, I'd suspect something wrong with your underlying sockets implementation. What system are you on? --Sean 19:25, 3 June 2008 (UTC)
P.S., the following C client will do the same thing as the Python one above. If it does not set your client interface correctly, your sockets implementation is borked:
#include <sys/types.h>
#include <sys/socket.h>
#include <netinet/in.h>
#include <arpa/inet.h>
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>

#define die() do { perror("error"); abort(); } while (0)

int main()
    char *client_ip = "";
    char *server_ip = "";

    int fd = socket(PF_INET, SOCK_STREAM, 0);
    if (fd < 0)

    struct sockaddr_in server = { .sin_family = AF_INET,
                                  .sin_port = htons(9000) };

    if (inet_pton(AF_INET, server_ip, (void *)&server.sin_addr) < 1)

    struct sockaddr_in client = { .sin_family = AF_INET,
                                  .sin_port = htons(0) };

    if (inet_pton(AF_INET, client_ip, (void *)&client.sin_addr) < 1)

    if (bind(fd, (const struct sockaddr*)&client, sizeof(client)) != 0)

    if (connect(fd, (struct sockaddr*)&server, sizeof(server)) != 0)
    return 0;


Why doesn't Amazon have an Australian version? Interactive Fiction Expert/Talk to me 11:49, 3 June 2008 (UTC)

maybe they are waiting for a strategic partner like you who sees long-term value in the online merchant beyond the fact that Amazon has not been able to generate a lot of profit since the dot com bust. No kidding. (talk) 13:42, 3 June 2008 (UTC)
Your comment doesn't make sense. was unprofitable during the dot-com boom, and only became profitable after the bust. -- Coneslayer (talk) 14:45, 3 June 2008 (UTC) is profitable? Sign me up for 100 shares! Just kidding,'s cumulative profits continue to remain negative. There is a lot of long-term value in Amazon, I agree. Many people idolize its customer service and Amazon Prime is a great idea. I also agree that the future looks bright for Amazon. Amazon Prime in Australia would probably make as much sense as it would in a small town in Wyoming. I don't care what happened during the dot com boom. Anything that happens in a crazy boom, stays in the crazy boom. :) Kushal (talk) 17:25, 3 June 2008 (UTC)
I think I'm using "profitable" in the most common way, considering a year (or a quarter, etc.) at a time. This is how it's used in our article, and in this sense, Amazon is profitable, and General Motors is not. While you certainly can integrate over the lifetime of the company, I don't think that's how most people define profit or loss. The impact of Amazon's past losses would be reflected in its outstanding debt. -- Coneslayer (talk) 19:26, 3 June 2008 (UTC)
Coneslayer, I think we can agree that in a few years, Amazon could be "profitable" from either point of view. Kushal (talk) 23:28, 3 June 2008 (UTC)

I can fully appreciate why you would like to see an Aussie Amazon; the postage from the UK or USA is quite high. My friends in Oz, will order books, etc and have them sent to my place here in the UK. I them post them on, in exchange for them getting me something from home (eg boxes of Cherry Ripes) or often Australian releases of DVDs are way ahead of those that come out in the UK and are often superior. I have obtained a full set of CountDown DVDs this way.-- (talk) 19:28, 3 June 2008 (UTC)

Does that mean overcharges for shipping for its downunder customers? 80, is it cheaper for your friends to have you ship it to them instead of have Amazon ship it directly? I thought it was something that only the cheapskates at the auction website did that. If this is true, it is very sad. I thought tried to do things right. :( Kushal (talk) 23:26, 3 June 2008 (UTC)
It may be that by going through a friend you avoid declaring the item's proper value or avoid paying tariffs in some other way. APL (talk) 01:38, 4 June 2008 (UTC)
You will always get cheaper postage from to the UK than to Australia. Sometimes I can post items to Australia cheaper than amazon. For instance, it you purchase a talking book, amazon will charge you a book rate rather than a CD rate. Also, I can mark the item "gift" and wrap it in pretty paper which removes tariffs. Finally, I can get things from home (Oz) that I can't get here in the UK. Everyone wins, apart from the taxman! -- (talk) 23:14, 4 June 2008 (UTC)
I thought book rates are cheap. Huh. I think I need to look at the postal service's list of fees. Kushal (talk) 00:22, 5 June 2008 (UTC)

Matlab Question ![edit]

In Matlab How to save and retrieve a multidimensional array from a text file? -- (talk) 13:51, 3 June 2008 (UTC)

According to this and this, it would be something like:
multi_array = cat(3, [2 8; 0 5], [1 3; 7 9])
save my_data.out multi_array -ASCII
to save it, and:
multi_array_again = load('my_data.out','-ascii')
to load it back in. --Sean 14:33, 3 June 2008 (UTC)

When i tried what you gave i got the following error -

Warning: Attempt to write an unsupported data type to an ASCII file. Variable 'multi_array' not written to file

I tried it without -ascii i got the following error while loading the file.

??? Error using ==> load Number of columns on line 2 of ASCII file <filename> must be the same as previous lines. —Preceding unsigned comment added by (talk) 06:35, 6 June 2008 (UTC)

Yahoo mail[edit]

I use Yahoo mail, and also subscribe to the en.wp mailing list. The problem is, the mailing list is ass-trociously huge and I get ~7 e-mails a day from the list. I usually end up moving them all to a designated folder, but surely there must be an easier way to do this. Can I automatically move all e-mails with subject X or sender X to a certain folder? Ziggy Sawdust 16:36, 3 June 2008 (UTC)

Yahoo! Mail has a filtering feature that does what you want. In Yahoo! Mail Classic, filter configuration is under Options -> Mail Options -> Management -> Filters. -- (talk) 17:05, 3 June 2008 (UTC)
The thing is, I tried that before and then all wikien-l messages didn't get through at all. Ziggy Sawdust 18:01, 3 June 2008 (UTC)
Maybe, just to give it a try, please add a new filtering criterion. Maybe the subject line? Maybe the sender domain? Do you have Yahoo! Plus? Kushal (talk) 23:22, 3 June 2008 (UTC)
You said that no WikiEN-l message got through at all. What filtering rule did you use? -- (talk) 12:45, 6 June 2008 (UTC)

Populous: The Beginning demo install error[edit]

I recently tried to install the above program and got the following message: "An error occurred during the move data process: -1" I googled the message, and came up with a bunch of different sites discussing similar errors, only every site listed a three digit number, instead of a just -1. I checked my temp files and hard drive space as suggested, and both have room to spare. Any help? Thanks in advance for your efforts. You may reply here or on my talk page. --AtTheAbyss (talk) 16:59, 3 June 2008 (UTC)

The number is usually an error code, that the program spews out if something goes wrong. The programmer can then debug by looking up what the code means. (e.g. ran out of hard disk space, etc) -1 sounds like a generic 'unknown error', though, so it isn't helpful. I can't really think of any advice than the usual. Run a scandisk? Make sure everything else is switched off? Check the place you are installing to, and delete anything left in it?--Fangz (talk) 22:47, 3 June 2008 (UTC)
I've tried the above ideas to no avail. Thanks for the effort.-- (talk) 04:09, 5 June 2008 (UTC)

IP address of sender in google group[edit]

Hi, I am a part of a google group.. Is there a way to trace the ip address of another member of google group who has replied to a posted message on google group?? If yes , how?? Its important for me to know somehow..... —Preceding unsigned comment added by (talk) 17:24, 3 June 2008 (UTC)

No. You don't get the actual sender's IP info in the message. At most, you get the email address the sender is claiming to be using - which may or may not be true. Google purposely masks that information and you'll have to take Google to court to force them to hand over the IP address. Even then, it will likely be a dynamic IP that doesn't belong to the user anymore after you go through all that trouble. -- kainaw 19:40, 3 June 2008 (UTC)
If I understand correctly, then many "Google groups" are just mirrors of usenet. If someone has posted to usenet directly, the message headers may contain some usefull information, if not an exact IP address. APL (talk) 19:44, 3 June 2008 (UTC)
even then, what if the user in question is using proxy or something like Tor? If you think the person is genuine, please write to him/her/them/it directly. If you suspect malice, expect that they have taken steps to hide themselves from you. Kainaw, if Google were to give up personally identifiable information without a fight, we would not have a template that lets us search Google like {{Google|this}}. Kushal (talk) 23:14, 3 June 2008 (UTC)

AT&T early termination[edit]

If I have a wireless plan with AT&T for about six months and I have to cancel my contract, I will need to pay the early termination fee and any monthly fee due. However, do I need to return my cell phone? I am pretty sure I don't have to. Please let me know if you have been in such a situation. Kushal (talk) 17:58, 3 June 2008 (UTC)

That depends on what kind of an agreement you have made with AT&T. Did you lease the phone from them? What does it say in the paper you presumably signed when you signed up for the plan? Really, all you need to do is call up their customer service and ask them, they can undoubtedly tell you how it works. -- Captain Disdain (talk) 18:17, 3 June 2008 (UTC)
Thanks a lot. I dialed 6-1-1. It appears that all I need to do is to call them on the day I want to have the service canceled and tell them to IMMEDIATELY cancel the contract. The representative said that my contract would then be canceled. I would not need to return anything. My bill would show the early termination fee and the prorated monthly fees. Kushal (talk) 18:29, 3 June 2008 (UTC)

Versace typeface[edit]

What typeface is the word "VERSACE" in this image written in? Also could you please give me instructions on how to download the typeface and apply it to Windows Live Messenger --Hadseys 21:32, 3 June 2008 (UTC)

It's Radiant URW T Bold. I'm pretty sure that fonts applied to Windows Live Messenger conversations only show up on the other end if the other person has it installed too (so if the other person doesn't have Radiant they'll see Arial or some other font). — Matt Eason (Talk &#149; Contribs) 15:37, 4 June 2008 (UTC)
(could also be Radiant RR Bold - they're very similar faces) — Matt Eason (Talk &#149; Contribs) 15:39, 4 June 2008 (UTC)
Doesn't seem like it's Radian RR Bold. If you look closely at the letter "E" in that font, you'll notice that the horizontal lines are slightly tapered. They don't seem to be in the image the OP pointed to. -- (talk) 13:01, 6 June 2008 (UTC)
Well spotted. It's probably Radiant URW then. — Matt Eason (Talk &#149; Contribs) 13:41, 6 June 2008 (UTC)

iTunes: flag mp3 as audiobook?[edit]

Is there a way to tell iTunes 7 that an arbitrary mp3 file is an audiobook and should be classified in the library as such? I've at least tagged it as "remember playback position" and have it filed as "books and spoken", but iTunes still sorts it as music. Only Google references I've found have been for legacy versions of iTunes. — Lomn 22:39, 3 June 2008 (UTC)

Well, for MP4 files, you could change the extension to .m4b (the default is .m4a) and re-add them to library. But MP3 files are of different format, so renaming them to .m4b probably won't work. But you can convert your MP3s to MP4. --grawity 12:04, 4 June 2008 (UTC)

Google Pages[edit]

Hello. Can a website master implement RSS Feeds into a Google Page website? Thanks in advance. --Mayfare (talk) 23:52, 3 June 2008 (UTC)

I don't know why you cannot. You have a fairly good access to the underlying html code. I think it is very possible. I need to do more research to find out how to do it, though. Kushal (talk) 00:18, 5 June 2008 (UTC)