Wikipedia:Reference desk/Archives/Computing/2008 September 1

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Computing desk
< August 31 << Aug | September | Oct >> September 2 >
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.

September 1[edit]

Downloading from[edit]

Hi folks. I have a major problem downloading from mybloop. Apparently all the files i download seems to give me an error "Access denied. This file cannot be shared in this way. Please use the links given on " when using FireFox. When i use Internet Explorer, it says the page could not be found. Even download accelerators cannot connect to MyBloop. The weird thing is that my friends can download everything they want on MyBloop. How do I solve this problem? Thanks in advance 1st September —Preceding unsigned comment added by (talk) 02:59, 1 September 2008 (UTC)

Well really you should ask them but it sounds to me that when it says "use the links" it is saying, "don't link directly to the file", if that's what you are doing. Are you getting to the file through the site's navigation or are you trying to download it directly? Many sites don't let you download files directly because that causes you to miss all the ads that keep the site afloat. -- (talk) 05:24, 1 September 2008 (UTC)


Is there any software in existence which can tell the owner of a website when and who has downloaded an image from their page? Is it available for a normal mortal like me to integrate into a website? --S.dedalus (talk) 03:01, 1 September 2008 (UTC)

Server logs have that sort of information in them. If you don't have an ability to access/use those, the only other way is to serve the images up with some sort of script that'll take down that information. It could be done in PHP, for example. It would add some additional work for the server to do but if well-written it wouldn't be a problem.-- (talk) 04:19, 1 September 2008 (UTC)
Even server logs can't tell you whether someone merely VIEWED your photo - or whether they DOWNLOADED it - because those are identical operations as far as your server is concerned. If you are concerned with the possibility that someone is violating your copyright on the image...forget it...there is absolutely no way to tell - and no way to prevent it. SteveBaker (talk) 21:22, 1 September 2008 (UTC)
Well, you can enforce your copyright through the normal mechanisms. But merely downloading something that you've uploaded for viewing is not a copyright violation in any way. (As noted, downloaded and viewing are the same thing from a technical point of view.) -- (talk) 00:43, 2 September 2008 (UTC)

Thanks! That’s exactly what I needed to know. Is there any way of simply making an image undownloadable? --S.dedalus (talk) 05:57, 2 September 2008 (UTC)

Not really - the user has to download it to their computer in order to view it, and once there they can potentially copy it (the best you can do is make it difficult for them to do so, but at worst they can always take a screenshot and get it that way). The best solution is not to put pictures online if you want to protect them, but if it needs to go online you should only add low-resolution versions so that the applications are limited if someone does try to reuse them, or watermark the images. - Bilby (talk) 06:11, 2 September 2008 (UTC)

Super High-Rez Maps.Google imagery?[edit]

While looking up cow pictures to try to witness the cow-compass phenomina for myself, I noticed this page [1]. What's the story behind this area? It looks like those high-rez photos are from a REAL low altitude. I notice that some of these patches are roughly in a straight line, but otherwise it seems completely random. Why and how does google have patches of super-high-rez images for bits of land smaller than a football field, in the middle of a sea of very poor low-res images? It seems like an interesting oddity. It makes me feel like I should go up to the roof of my building with my camera and take a photo of the parking lot for Google. APL (talk) 05:32, 1 September 2008 (UTC)

I think I read somewhere that these were taken by the National Geographic for a feature or some such. fake edit: Aha found it - [2] I was right (talk) 08:27, 1 September 2008 (UTC)

Neat! Thanks. APL (talk) 17:10, 1 September 2008 (UTC)

c programming[edit]

Hi there any way by which I can use the functions created in one C program in another program without rewriting the fuctions all over again?? i.e. I have some functions in say "file1.c" which I would like to use in "file2.c" how can I do this?? This must be similar to including header files in a program but how do I include the whole program?? Please help...

Thanking u in advance... —Preceding unsigned comment added by Piyushbehera25 (talkcontribs) 10:16, 1 September 2008 (UTC)

While I have never programmed in C myself I believe you should be able to either include the shared functions in something like a header file, or compile the functions into a library and link them to the different projects you have like other C libraries, or just have multiple source code files and link them together at compile time in a modular fashion. Phydaux (talk) 11:13, 1 September 2008 (UTC)
The way to do this is to create your own header file. Put function prototypes for the functions you want to share into a file with a .h extension. The name doesn't matter, but it's probably best to name it file1.h so that it's obvious where the function bodies are. Second, include that header file into file2.c with a line like #include "file1.h". (Don't use #include <file1.h> since that tells the compiler to look for a system header file by that name.) Third, tell the compiler to compile both file1.c and file2.c and link them together. If you're using an IDE you can do this by adding both .c files to the project. If you're using a command-line compiler you can just list both .c files on the command line. If you want to be more sophisticated you can create a Makefile, but for a small project with just two .c files there's probably no need. -- BenRG (talk) 15:02, 1 September 2008 (UTC)



Using Google, how can I find results of a word or phrase that are not on a particular site? For example, results for "Mario" on any site except Wikipedia. February 15, 2009 (talk) 11:07, 1 September 2008 (UTC)

Easy, use
Using your example you would use: Mario
Phydaux (talk) 11:16, 1 September 2008 (UTC)
The minus sign can also help you find sites without that word, as well. So if you wanted to look for Mario, but exclude all results with "Nintendo", you'd do Mario -Nintendo. --Alinnisawest,Dalek Empress (extermination requests here) 14:10, 1 September 2008 (UTC)
Some advanced search tips from Google and a cheat sheet - Phydaux (talk) 14:44, 1 September 2008 (UTC)

Keyboard Locks Up[edit]

Every once and a while, my MacBook keyboard will strangely lock up. I've only experienced this problem while using Firefox, but every time it occurs, I have to restart my MacBook for the keyboard to respond. I'd normally think that it's just a hardware error, but my mouse is always completely responsive. So, I'm just wondering if (1), anyone has ever experienced this, and (2), does anyone know a way to fix the problem without having to completely restart the comp? Mastrchf (t/c) 15:42, 1 September 2008 (UTC)

1) I have an (intel) mac and never experienced it. Even on Firefox, my main web browser. 2) Have you tried just restarting firefox, or a wireless keyboard? (talk) 18:18, 1 September 2008 (UTC)

  • Can you work if you plug an external keyboard? NerdyNSK (talk) 14:46, 6 September 2008 (UTC)
I'll try an external keyboard next time it locks up. Thanks for the help guys. Mastrchf (t/c) 16:30, 6 September 2008 (UTC)

Data structures[edit]

What is 'Abstract data types'? —Preceding unsigned comment added by (talk) 17:22, 1 September 2008 (UTC)

It would be so great if there was... like... a huge encyclopedia online, for free, where you could look this stuff up! (talk) 17:26, 1 September 2008 (UTC)
...Or if classes had accompanying textbooks with indexes in the back to quickly look up the exact answer the teacher is looking for. -- kainaw 17:39, 1 September 2008 (UTC)
The textbook might be a better bet; our article on abstract data types is spectacularly bad. rspeer / ɹəədsɹ 17:58, 1 September 2008 (UTC)
It's better now—I rolled it back to an earlier, more coherent version. -- BenRG (talk) 20:14, 1 September 2008 (UTC)
No but seriously, I want to know too. I've been re-reading Code Complete and there's stuff in there about it but I still wasn't quite sure. ADT != class? ADT subseteq class? —Preceding unsigned comment added by (talk) 20:15, 1 September 2008 (UTC)
"class" is a keyword used in some languages... "abstract data type" is a concept that can apply in any language that has types. For a good example of abstractness, see the fpos_t type in C's <stdio.h>. It represents a position in a file. In most cases it's an integer, a simple count of how many bytes you are from the beginning of the file. But you can't assume it's always an integer. The specification is loose enough to allow it to be a struct containing multiple integers, or a pointer, or even some kind of floating point monstrosity. The only way to get a valid fpos_t is by calling fgetpos(), and the only legal thing you can do with it after you got it is call fsetpos() to return to the file position where you were before. So it's very abstract, but is it a "class"? C doesn't have classes. --tcsetattr (talk / contribs) 20:46, 1 September 2008 (UTC)
From the responses so far, and from the article, it appears that abstract data type may mean different things depending on which language you are referring to. The example given by tcsetattr is simply a typedef for a long or a longlong / __int64. Closer to the spirit of what tcsetattr alludes to is the concept of an opaque pointer. However, I don't think it is common to refer to neither typedefs nor opaque objects as 'abstract data types'.
In C++, an abstract data type is a class, which specifies an interface while deliberately leaving the implementation unspecified. All its functions are pure virtual functions, and it has a virtual destructor. You use it by inheriting from it. When doing so, you essentially promise that your class will comply with the "contract" that is specified by the class declaration. The benefits are twofold: (1) you can write generic code, which will work with any object that complies with the interface; (2) you will minimize interdependencies in your code - clients depend only on the abstract data type, not on a particular implementation. This will also increase compilation speed, as the abstract data type header file will typically be a lot smaller, and have fewer, if any, include files itself, than the header file of a typical implementation. Using abstract data types is also a means of breaking cyclic dependencies. --NorwegianBlue talk 18:09, 2 September 2008 (UTC)
Are you sure that what you describe matches some definition of an "abstract data type"? It seems to me that what you're describing is an abstract class, a somewhat narrower concept. Confusingly, the latter article redirects to "abstract type" on wikipedia, but there is still a distinction to an "abstract data type". In any case, it seems to me that software engineering concepts are generally defined in multiple conflicting ways in different communities. (talk) 18:21, 2 September 2008 (UTC)
No, I'm not aware of any official definition of ADT, although the article contains this link, which I find somewhat self-contradictory (it contains data, but is independent of the implementation). Bjarne Stroustrup uses the terms "abstract type" and "abstract class" interchangably in "The C++ programming language (third edition)". I limited my answer to C++, because, in that context, the terms are synonymous AFAIK. --NorwegianBlue talk 19:14, 2 September 2008 (UTC)
I picked fpos_t specifically because you don't even know whether it's a pointer or not. The "typedefs don't count argument" is just C++ bigotry. "What? C can't have all our cool features without the extra language complexity! It must not be real abstractness." Baloney. What's more abstract than a type that you can't do anything with except call a get and a set function? --tcsetattr (talk / contribs) 20:00, 2 September 2008 (UTC)

Data structures[edit]

PDLC-Program development life cycle —Preceding unsigned comment added by (talk) 17:25, 1 September 2008 (UTC)

Yes? What about it? --Alinnisawest,Dalek Empress (extermination requests here) 17:49, 1 September 2008 (UTC)
Silly me, I've been looking all over for Polymer-Dispersed Liquid Crystals lol. The main stages of the program development life cycle are: development -> testing -> market introduction -> growth -> maturity -> saturation -> decline. The Product life cycle management article goes into each stage in more detail. JessicaThunderbolt 20:12, 1 September 2008 (UTC)


I can I run two Firefoxs at the same time. When I open one then when I open the next it says it can't cause one is already running. I want to run two at same time. How? Thank you in advanced very much. (talk) 18:56, 1 September 2008 (UTC)

Did you try File > New Window? Admiral Norton (talk) 19:33, 1 September 2008 (UTC)
If you want to run multiple Firefox profiles, say for example if you share a computer with roommates or something, you can set them up using profilemanager. This page explains how, and this one gives some more complicated methods. If you want to log into the same site under multiple different accounts, an easier option is the cookiepie extension which creates a set of cookies for each different tab or window simultaneously. JessicaThunderbolt 19:58, 1 September 2008 (UTC)
Mozilla Prism may also be of interest to you. Kushal (talk) 22:44, 2 September 2008 (UTC)

New Keyboard[edit]

I just got a new emachines keyboard and it has special buttons, the regular keyboard is working great, but the special buttons are unresponsive. I have an Insignia. The Keyboard model # is KB-0108. How do I get them to work?Elatanatari (talk) 20:08, 1 September 2008 (UTC)

You need to download and install a driver. Looks like this is the one you'll need. JessicaThunderbolt 21:57, 1 September 2008 (UTC)

IT qualifications[edit]

In the United States (specifically Texas), must one possess a license, certification, etc., in order to legally repair computers for a profit? Ζρς ι'β' ¡hábleme! 23:33, 1 September 2008 (UTC)

A private detective license, in case they find child porn and need to document it. Yes, Texas is batshit insane. I didn't know they went through with it though. --mboverload@ 23:34, 1 September 2008 (UTC)
Seriously? Ζρς ι'β' ¡hábleme! 23:37, 1 September 2008 (UTC)
Yes. Batshit Insane --mboverload@ 23:38, 1 September 2008 (UTC)
Agreed. That's freaking crazy! Ζρς ι'β' ¡hábleme! 23:45, 1 September 2008 (UTC)
I just skimmed through the law and didn't see anything about regular computer work. This story seems to say that it's for technicians performing forensic work and investigating the actions of a "third party." Check the discussion.--Birdsusing nnn (talk) 11:21, 2 September 2008 (UTC)