Computing desk
< November 1	<< Oct \| November \| Dec >>	November 3 >

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.

November 2[edit]

Facebook Privacy[edit]

A website "connected" my facebook account to my account on their site because I accidently clicked "connect facebook" a year after I made the account and there is no way to unconnect it via there site, I blocked their app on facebook but my account is still connected. Is there anyway to fix this besides deleting my facebook? I contacted the site but they are no help. I have been on their site a long time and don't want to make another account. --Wrk678 (talk) 00:02, 2 November 2012 (UTC)[reply]

You are not alone, I have the same issue and have found no solution through either site. μηδείς (talk) 00:23, 2 November 2012 (UTC)[reply]

Well, I really need to unconnect it, shirley there must be a way. Can anyone help?--Wrk678 (talk) 13:32, 2 November 2012 (UTC)[reply]

If the site refuses to help you, you're stuffed. Basically you've done all you can do on Facebook. It's up to the site to let you unconnect your profile on their site from Facebook. If they won't let you, reconsider whether the site is worth it. Nil Einne (talk) 16:26, 2 November 2012 (UTC)[reply]

Since FB doesn't seem to roll out the same changes to all their users all at once (eg I know some people still don't have the Timeline theme to their account), I'm unsure if this will work but... Go to the upper right hand corner of the FB page and click next to your name to bring down the pull down menu. In it you should find an option for Privacy Settings, go to Ads, Apps, & Websites. Check if the web site is listed there. Dismas|^(talk) 17:53, 2 November 2012 (UTC)[reply]

I changed that setting, but it didn't help. Any other ideas? --Wrk678 (talk) 15:40, 3 November 2012 (UTC)[reply]

Source code to create a site like The Pirate Bay[edit]

Does anyone know where I could get a source code that could help me create a site like The Pirate Bay — Preceding unsigned comment added by 189.220.70.192 (talk) 05:20, 2 November 2012 (UTC)[reply]

I can't look into it now because I am at work, but I thought several months ago they announced that by switching over to tracker-only the entire site was reduced to a few dozen MB. The idea was to make it easy to distribute everything needed to run a copy of the site, so that new copies could pop up to get around blocks. Our article doesn't mention anything about it, but you should look around the pirate bay site and see if you can simply download a copy. Be careful - torrent sites tend to have to deal with a lot of legal trouble. I know we can't offer legal advice, but it is just a very bad idea to start something like TPB without a very detailed knowledge of how the laws in your region will protect you. If you're only hosting torrents to things you have the rights to you should be ok. 209.131.76.183 (talk) 12:31, 2 November 2012 (UTC)[reply]

git clone git://git.what.cd/gazelle ¦ Reisio (talk) 19:25, 2 November 2012 (UTC)[reply]

@Reisio What's that? What is it for? How can I use it? — Preceding unsigned comment added by 189.220.70.192 (talk) 05:26, 3 November 2012 (UTC)[reply]

Read the git article, run the command, read the docs. ¦ Reisio (talk) 15:54, 3 November 2012 (UTC)[reply]

http://www.xbtitfm.com/ 77.166.70.218 (talk) 15:21, 5 November 2012 (UTC)[reply]

Browser memory and Wikipedia's saved edit summary[edit]

This is more a computer (browser) question than a Wikipedia one, that's why I am asking here.

Short version

Is there any way to quickly delete all saved edit summaries other than deleting browser history?

Long version

(You don't need to read long version if you have already got what i am asking)

In Wikipedia following WP:FIES I write edit summary for every edit. Generally in every 4-5 days I delete all saved edit summary. I can not delete all browser history and cookies because I need those. I manually delete edit summaries by hovering cursor on one and pressing Del in keyboard.

But, the problem is in 4-5 days when I delete those saved edit summaries, sometimes I have 400, 500 or more entries there. Now pressing the Del button 500 times continuously is a tiring work. Last month I had some 5000+ edits in English Wikipedia and some edits in Commons and other Wikipedia too. So I had to press the delete button more or less 5000 times.

Is there any way to delete all saved edit summaries? I tried pressing Ctrl A, but that does not work. --Tito Dutta (talk) 05:47, 2 November 2012 (UTC)[reply]

Try this: Go to edit any article. Erase whatever is in the edit summary box. While still in the edit summary box, press down on the keyboard to move to the top edit summary. Hold Shift and press Delete. The list of your edit summaries should stay open (in Firefox at least - I can't really attest for other browsers as I don't use them). Now, holding Shift, press and hold Delete to remove all of your edit summaries quickly. -- 143.85.199.242 (talk) 18:03, 2 November 2012 (UTC)[reply]

Hello (Seo is best for girls ?)[edit]

Seo is best for girls or not?

Thanks and regards leela bisht←<ref>delhi</ref>¤ — Preceding unsigned comment added by 119.82.92.150 (talk) 14:29, 2 November 2012 (UTC)[reply]

I fixed your noninformative title and reformatted your Q, although I have no idea what it means. StuRat (talk) 16:21, 2 November 2012 (UTC) [reply]

What is "Seo" ? Site enhancement oil ? StuRat (talk) 16:21, 2 November 2012 (UTC)[reply]

(EC) It would greatly help if you'd mention what Seo is. I'm guessing you don't mean Search engine optimisation or Site enhancement oil? And if you do, what exactly do you mean. For example, are you wondering if search engine optimisation is a good career choice or if you should pay a SEO company to increase your rankings on search engine. Note that you question also sounds like a request for opinions which as the header says we don't do. Nil Einne (talk) 16:23, 2 November 2012 (UTC)[reply]

Optional return values in C++[edit]

In C, one frequently writes a procedure of the form

void calculation(const double * input1, const double * input2, size_t kt, double * ret1, double * ret2, double * ret3)

where the procedure calculates various things, but returns information by:

if (ret1) *ret1 = ...

if (ret2) *ret2 = ...

if (ret3) *ret3 = ...

Is there a legitimate way to do that in C++? — Arthur Rubin (talk) 18:08, 2 November 2012 (UTC)[reply]

You can still use the pointers. C++ also adds proper pass by reference:

#include <iostream>

void swap(double& a, double& b){
  double tmp = a;
  a=b;
  b=tmp;
}

int main(){
  double x=10.66, y=13.14;
  std::cout << x << "," << y << std::endl;
  swap(x,y);
  std::cout << x << "," << y << std::endl;
}

-- Finlay McWalterჷTalk 18:19, 2 November 2012 (UTC)[reply]

Either use overloaded functions or use a pointer just like in C. It is often possible to pass NULL for a reference variable but that is undefined behaviour, pointers like C are the thing to use. Dmcq (talk) 18:22, 2 November 2012 (UTC)[reply]

You might also think if there is an object-oriented solution. If those inputs and outputs are a "thing" that you can name, make the thing a class that has the inputs and outputs as fields, and calculate() is a method in the class. Then the caller can pick whichever of the outputs it needs. 88.112.36.91 (talk) 15:01, 3 November 2012 (UTC)[reply]

That last is not an acceptable solution; there are a lot of complex calculations, and I only want to perform the onces that will be needed. Basically, I will need to compute one or more of the slope, intercept, or standard error for a linear fit. The two calls I'm sure about are intercept and standard error; and intercept and slope, but I want the package to be able to compute any of them. It might work if the class "remembers" which of the calculations have already been done. — Arthur Rubin (talk) 17:31, 3 November 2012 (UTC)[reply]

The interface you're suggesting is pretty ugly. C++ style generally prefers references to pointers for the very purpose of avoiding creating null pointers (please do click that link and see the "billion-dollar mistake" story). Unless you're absolutely trying to bum single instructions, look for a cleaner interface. One obvious one is supply different entry points for the slope, intercept, etc. You could have a class that remembers which ones have already been computed. An option type is a clean way to implement such a thing. Boost (C++ libraries) have an option type here which is pretty trivial to implement yourself if you don't want to pull in the Boost stuff. 67.119.3.105 (talk) 15:12, 4 November 2012 (UTC)[reply]