Wikipedia:Reference desk/Archives/Computing/2006 July 24

Humanities Science Mathematics Computing/IT Language Miscellaneous 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 at one of the pages linked to above.

< July 23 Computing/IT desk archive July 25 >

Torrent

Let's say I wanted to create a torrent, not to go on a big website like IsoHunt or The Pirate Bay, but just so a few people I personally know could get a link to it. How would I go about doing that? zafiroblue05 | Talk 01:48, 24 July 2006 (UTC)

Azureus has a built in tracker, so you can make your own torrents. I believe UTorrent does as well, although I have never used Utorrent. See here, and here for more information about Azureus. Debigulator 02:18, 24 July 2006 (UTC)

If it is just a few people why use a torrent? Jon513 20:26, 24 July 2006 (UTC)

If you personally know them, why use a torrent as the above person mentioned? --Proficient 11:54, 26 July 2006 (UTC)

Horizontal Scrolling in a browser window

When I search the internet for "horizontal scrolling" and "HTML" I get many pages saying to avoid horizontal scrolling. I agree, but....here is what I want to do:

• I need to use a display that can show 1280 pixels horizontal
• I need to use a browser to display two images side-by-side, each image has a width of 640 pixels.
• I have to use a browser running under Windows XP.
  I have no problem getting this to work with Macintosh OSX. In OSX I can slightly increase the width of the browser window until both images display side-by-side. However, using either Explorer or FireFox in Windows XP, the two images always are shown stacked vertically, not side-by-side. When I try to increase the width of a browser window by draging the window's side border, nothing happens.
  What is a simple way to get this to work in Windows XP? I do not mind horizontal scrolling.
  This is the HTML code I am using to display the two images:
  <img src="image1.jpg" border="0" height="480" width="640">
  

<img src="image2.jpg" border="0" height="480" width="640">
--JWSchmidt 14:34, 24 July 2006 (UTC)

Use a table: <table><tr><td><img src="image1.jpg" border="0" height="480" width="640"></td><td><img src="image2.jpg" border="0" height="480" width="640"></td></tr></table> --Kainaw ^(talk) 14:56, 24 July 2006 (UTC)

• Thanks! That works great. --JWSchmidt 15:35, 24 July 2006 (UTC)

annoying : my azerty keyboard works as qwerty on foreign sites,like these

Hello, I am experiencing an annoying problem, and thank god for the recent creation of this desk.

In Belgium and France we use Azerty keyboards instead of Qwerty. Usually there is no problem, but today I went to IMDB, www.mathlinks.ro and Wikipedia itself, and everything is interpreted as Qwerty : when I press the 'a' button on my keyboard it gives a 'q' on screen. I went to the website www.vrtnieuws.be, which is Belgian, and I had no problems at all.

I have had this problem before but I think this is the first time I have it on all foreign sites. It is worth nothing that the last time it just seemed to disappear the next day.

So what do you think?

Evilbu 15:51, 24 July 2006 (UTC)

This is a tricky one. I would check that everything looks okay under 'Keyboard' and 'Regional and Language Options' in the Control Panel. What browser are you using ? Robmods 16:41, 24 July 2006 (UTC)

the quick brown fox jumps over the lazy dog : there my keyboard works perfectly fine now! I do not understand. I use Internet Explorer, my language is Dutch. Evilbu 17:47, 24 July 2006 (UTC)

Computers

When was the first computer built?

Did you check the article computer? Did you see the link to history of computing? --Kainaw ^(talk) 17:32, 24 July 2006 (UTC)

Actually, History of computing hardware would probably be more useful in this case. --ColourBurst 17:40, 24 July 2006 (UTC)

Depends on what you mean by "computer". --Serie 22:59, 24 July 2006 (UTC)

The ENIAC was the first wide-scale electronic computer, which is probably what the person asking the question was looking for, but I could be wrong. --Proficient 11:56, 26 July 2006 (UTC)

That's debatable, but ENIAC was the first American computer, and we all know that's all that counts... :P — Haeleth Talk 21:28, 28 July 2006 (UTC)

During World War II a "computer" was a person, most likely a woman, and was definitely not a machine. It would be impolite to say they were "built". Devices for computation might include the abacus, the Antikythera mechanism, or even Newgrange. Probably what you want is the article on the history of computing hardware. --KSmrq^T 18:22, 26 July 2006 (UTC)

quotation marks

Why are some windows' registry commands (data) enclosed in quotation marks?

If the data is a test string with one or more spaces in it, the quotes may serve to mark off the beginning and the end, otherwise the spaces may confuse things. Also, even if there are no spaces, some programs may write the data as text strings instead of numberic values. I'm not sure if that explains all the uses of quotes though... Digfarenough 20:19, 24 July 2006 (UTC)

In programming especially, strings are often enclosed in quotation marks to differenciate from variables. For example if you wanted to use the phrase "Tigers are big", but you already had three variables names, tigers, are, and big, which would the programming language think you were wanting to use? In Visual Basic, strings and phrases are enclosed in quotes to distinguish from variables. Harryboyles 22:30, 24 July 2006 (UTC)

All that aside, in the Windows registry, setting a value in quotation marks means disabling it. This is most commonly used in the section of the registry which controls what programs run at startup - by enclosing the program in question in quotation marks, you are preventing it from running at startup, but the value is still there so that you can easily re-enable it. — QuantumEleven 07:28, 25 July 2006 (UTC)

You learn something new every day. :-) Harryboyles 07:56, 25 July 2006 (UTC)

Firefox doesn't work with YTMND

and other sites with music/videos either. The bar on the top comes out and says I need additional plugins to view this page, then it says to manually install Quicktime, but I alredy have both Quicktime and WMP. It works fine in IE, though, and I have no idea what causes this Wizrdwarts (T|C|E) 18:23, 24 July 2006 (UTC)

If you had QuickTime installed before you installed Firefox, you'll probably need to re-install QT so that the proper plug-ins get put in the right places. You might get some help starting here. --LarryMac 19:41, 24 July 2006 (UTC)

Install Plugins. --Proficient 11:56, 26 July 2006 (UTC)

Proficient: I already installed them, but thanks anyways!

LarryMac: Thanks! It works now! Wizrdwarts (T|C|E) 23:14, 27 July 2006 (UTC)

Complexity...

Is it possible (in theory) for something to design something more complex than itself? Could I build a computer that would tell me how to program a computer to do something? If that is not clear, an example is:

1. I want to do task X.
2. I build a computer, A.
3. Computer A is not complex enough to complete X.
4. Computer A is complex enough to design computer B.
5. Computer B can solve X.

Is this banned by some fundamental rule of the universe/logic/computing/something else? Or would it be (theoretically) possible? —Daniel (‽) 19:54, 24 July 2006 (UTC)

I can't offhand think of any reason this couldn't happen. In fact, there's a fairly popular theory that that's how we'll eventually be rendered obsolete - intelligent machines will design more intelligent machines in an infinite cycle, and naturally they won't have any reason to keep us around. As a matter of fact, there's a fun book based on this premise. Black Carrot 20:04, 24 July 2006 (UTC)

Consider X = "compute 1,000,000 digits of π in a second", A = "human", B = "computer". Conscious 20:18, 24 July 2006 (UTC)

you built a human!? Jon513 20:22, 24 July 2006 (UTC)

It's simple. It just requires another human. --Zemyla^t 20:27, 24 July 2006 (UTC)

Of an opposite sex, you know. Conscious 20:37, 24 July 2006 (UTC)

I dunno, if Computer A is a universal computer (Turing equivalent), there ought to be no task that is too complex for it in theory. There are two other possibilities that come to mind, though: Computer A may be a finite state machine and the task may be to recognize a context-free grammar, so Computer A would not be able to perform the task (because it is not a universal computer).My guess (without proof) is that Computer A would not be able to build, e.g., a pushdown automata that could recognize the context free grammar. Another possibility is that Computer A has too little memory to perform the task (e.g. a task requiring that 2KB of values all be in memory simultaneously, whereas Computer A only has 1KB of memory). In that case, my intuition is that it would be possible for Computer A to design a Computer B that would be able to complete the task (assuming Computer A is universal). It's a good question though, I've pondered similar things in the past. Digfarenough 20:26, 24 July 2006 (UTC)

Yes. It is expected of AI programs to produce "logical" code that more complex than the AI code itself. For example, one of my undergad projects was to develop a distributed AI engine. That was very easy. Then, to test it, we first made a rating test to tell it if it was doing well at summing integers. Designing the rating test was difficult, but we got the hang of it. Then, we developed a test to tell it if it was doing well at sorting integers. After a year of fine-tuning the test, it developed a rather complex distributed sorting algorithm - though it wasn't the fastest one, it was complex (we never added speed to our test). Side note here: If you want a program to develop something very complex without explicity telling it how to do it, you need to provide some form of test that rates the end result. That test can be very difficult to develop - possibly more complex than the program being developed. --Kainaw ^(talk) 23:22, 24 July 2006 (UTC)

I think what you are referring to is called a genetic algorithm. It searches through a list of possible solutions to a given problem and combines them at random to produce new solutions. It then uses a special test (called a fitness function) to choose the best solutions. These solutions are htne allowed to be combined again. It has been proven that given a good fitness function, the algorithm will converge to an acceptable solution after a given number of generations. However, it is possible that it will converge to a local peak in the solution space rather than to the best solution. My MS thesis ivolved the use of genetic algorithms and the example I used found a maxima on a graph of a sine function. It convereged to a solution after about 6 generations.

Another example from AI would be the use of neural networks, which construct algorithms to solve a given problem based on some training data. The algorithms themselves usually are more complex than the program used to construct the neural network. --Max

Yes, that's impossible. Specifically here:

Computer A is not complex enough to complete X, yet is complex enough to design computer B which can complete X. That's not possible - if it could make a machine that can do X, it can also do X itself.

Be very careful when defining 'complex' and don't define it as speed. A human, if you spent enough time, can calculate 1 billion digits of Pi, yet normally you would call that impossible - it's not, it's just slow.

Look at the history of computing: have humans EVER created a machine that can do something that humans can't do? It's never happened before, and it never will. (And remember don't talk about speed, or silly things like fine motor skills or strength, I mean actual creative activities.)

The article on Technological singularity talks about this, but in my opinion does so incorrectly. It starts by assuming your conjecture is possible, without ever actually proving it, or even doing something as simple as suggesting how it might be done. I wrote a paragraph about the problem, but it was removed from the article since I didn't have a site-able sources. 71.199.123.24 21:46, 27 July 2006 (UTC)

You know, he didn't actually say the person couldn't do it, just that the first computer couldn't do it, but could design something that could. You have a good point, though, the criteria can be broken down much further. Speed is an issue, but that can be left out. Memory is another, since it puts a major upper limit on what size of problem it can tackle, but let's say that's not a part of "complexity". There are the things that get hooked into computers, sensory apparatus you might say, which are far more sensitive, more precise, and faster than ours, but again, not complexity. How about swarm intelligence and the like, which I've been reliably informed quickly far surpasses any human's ability to fully understand or, more importantly, mimic. However, if you really want to push a point, a human could technically work out every line of code and every string of arithmetic by hand, in order, and in a few billion years wind up with the same result. Even if it was split up into seperate processors, a feat no human brain is capable of, you could alternate between them and still theoretically eventually finish. So, to surpass ourselves in "complexity", it seems we would have to find at lest one line in the code that we were inherently incapable of working out ourselves. Quantum computers could provide that. Black Carrot 21:09, 28 July 2006 (UTC)

DVD+RW to Computer

How to you upload data from a DVD+RW to a computer, for a use in a program (such as Windows Movie Maker). Hello32020 22:55, 24 July 2006 (UTC)

Insert the DVD+RW disk into your computer's DVD drive (which needs to be +RW capable). View the disk contents as a window. Copy the contents to the folder of your choice on the hard drive. --Canley 23:46, 24 July 2006 (UTC)

Magical onClick="return (false);"

I used MS Word to make a few short HTML files. I saved the files as plain text and gave them the extension .html
These HTML files worked fine on my computer. In particular, they had working hypertext links such as:
<a href="Images Page 2.html">Next page of images</a>
I sent these files as email attachments to some other people. The same HTML files now on their computers , mysteriosly contain code for the links that looks like this:
<a href="Images Page 2.html" onClick="return (false);">Next page of images</a>
Question: where did the inserted onClick="return (false);" come from? How can I avoid this in the future? --JWSchmidt 22:57, 24 July 2006 (UTC)

MS Word added it. It is a word processor, not a web page design tool. So, you shouldn't expect it to produce useable HTML. --Kainaw ^(talk) 23:17, 24 July 2006 (UTC)

next time use Nvu. Jon513 23:35, 24 July 2006 (UTC)

• I have in the past used BBEdit Lite and Netscape Communicator. Thanks for the link to Nvu....I had no idea that it exists. --JWSchmidt 00:45, 25 July 2006 (UTC)