Wikipedia talk:FAQ/Technical/Archive 1

How big is the database?
The database is roughly 4 GB in size (including all languages and support tables) and growing.


 * Does this inculde the images? (images stored in the tables or as files?) Fantasy 15:06, 11 Aug 2003 (UTC)


 * Wow, that's soooo no longer up to date. :) Anyway, images are stored as files, separate from the database. --Brion 21:47, 11 Aug 2003 (UTC)

Salting passwords?
From this page: Wikipedia stores passwords as a salted MD5 hash. This means that if we have a security breach and someone reads our hard drive, it will generally be difficult for them to get your password, requiring weeks or months of computation time per password. The exception is if you use a password which is a word in any language, a person's name, a phonetic spelling of a word, etc. Then they may be able to break our encryption in minutes. However, from User talk:Tim Starling/Password matches: And no, the hashes aren't salted. -- Tim Starling 01:02, Jul 9, 2004 (UTC) From the above, and, indeed, the existance of a page matching users with identical passwords, it seems that Wikipedia does not store passwords as a salted MD5 hash, as claimed; something should be modified, either this page, or (preferably) the password hashing method (there's a real threat if the password hashes are ever stolen, e.g. http://passcracking.com/ ). &mdash; Matt 23:08, 25 Aug 2004 (UTC)


 * I wrote both of those quotes. At the time I wrote that feature, I did not have shell access, so I wasn't in a position to make sure it was implemented. I assumed it was, initially, but later found out that it wasn't.


 * Note that checking for matching passwords has been proposed as the primary means of merging accounts currently spread across various wikis. The properties of the non-salted hash have also been used from time to time: to copy user accounts, to synchronise passwords across wikis, and of course to check for sock puppets. There's an argument to be had for leaving it the way it is.


 * A more serious security problem is the possibility of an attacker obtaining a user's password hash via a cookie, when the user has the "remember password" feature enabled. It's entirely possible to allow persistent logins without sending out anything related to the password. -- Tim Starling 16:29, Aug 27, 2004 (UTC)


 * For the record, both of these things were later done on Wikimedia. -- Tim Starling 07:46, 20 June 2007 (UTC)

Problem pictures
Certain articles contain pictures that are causing problems. These pictures not only wont load, but also stop all other pictures on other pages from loading after the problem picture has failed to load.

There is one on Earth. (The second picture down I think). Now I have gone there and I can't load any pictures (even outside of Wikipedia). If I reboot I will be able to again. Bensaccount 19:59, 14 Sep 2004 (UTC)


 * I am not having the same problem. What's your browser and connection?  (Mine's Mozilla 1.7.2 on Win2K on an SDSL line behind a NAT/firewall.) - jredmond 20:10, 14 Sep 2004 (UTC)
 * Maybe a browser problem? I've never experienced anything like this (Firething), though I occasionally encounter an invisible image, or one that won't display (usually just a missing thumbnail); the Earth page all loads fine for me. -- Wapcaplet 20:15, 14 Sep 2004 (UTC)
 * That page is fine for me as well. Have you changed anything in your browser settings that could have caused it? Have you tried using a different browser? If it's an ongoing problem, please report it at . Angela. 21:54, Sep 14, 2004 (UTC)
 * I have a (unfortunately very vague) recollection of someone somewhere (I think here) stating some versions of MS-IE have a bug related to trying to display multiple JPEGs that sounded very similar to what you are describing. It also may be related to this. Either way, I believe the answer is to update IE to the latest version (for the security fixes, if nothing else), and/or switch to one of the solid IE alternatives, such as Opera or Mozilla. I am able to more or less duplicate the problem on an NT box using IE 5.0, but don't have any problems with more up-to-date configs or alternate browsers. Niteowlneils 23:53, 14 Sep 2004 (UTC)

Official Wikipedia search box?
Does Wikipedia offer an official toolbar that people can to search for an article from another site (along the lines of the Google Free web search or the Dictionary.com searchbox)?

Acegikmo1 23:36, 14 Sep 2004 (UTC)


 * Well, you may use Firefox. Next to Firefox's address bar, there is a customizable search bar that lets you add search engines. You can add Wikipedia too. -- Etz Haim 00:38, 15 Sep 2004 (UTC)


 * Given that Wikipedia often disables full-text search to keep the servers running, a toolbar search interface to Wikipedia isn't quite as useful as it could be. :-/ Good idea, though. And thanks for the Firefox tip; I'll have to try it out. &bull; Benc &bull; 01:02, 15 Sep 2004 (UTC)


 * One option is to add a bookmark in Mozilla (likely works in Firefox as well) to http://en.wikipedia.org/wiki/Special:Search?search=%s&go=Go and put w in the keyword field of the bookmark property (Bookmarks->Manage Bookmarks...). To go to an article in Wikipedia, type w Article Name in the address box of Mozilla to go directly to that article.  Replace fulltext=Search instead of go=Go in the bookmark URL to do full text searching, if available. Al guy 01:27, Sep 15, 2004 (UTC)


 * I think you don't want a search bar but HTML for a search box you can place on your own website. I don't know of any "standard" box, but here's one I just whipped up (just for you):




 * Perhaps others will have use for it as well. Derrick Coetzee 02:50, 19 Sep 2004 (UTC)
 * XHTMLized 20 Aug 2005

Add link to TomeRaider database?
It seems that in the section talking about how to download the Wikipedia database the TomeRaider_database should be mentioned. At least, it was what I was looking for. (Apologies if I haven't followed convention, it's my first time. I've read the FAQ, doing my best--for a "simple" system there sure seems to be a lot of rules!). Rich Fletcher 20:14, 27 Dec 2004 (UTC)

Is question 9 needed anymore?
Question 9 reads Why are some of the pages scrunched together and capitalized LikeThis?. I have been here since 2003 and have never encountered a camelcase titled article. The existence of this question seems to be a historical anomoly. Is there any opposition to removing this question, either entirely or to a no-longer frequently asked questions page? Nohat 18:16, 22 Mar 2005 (UTC)


 * After several weeks with no comment, I have removed this question. Nohat 14:27, 11 Apr 2005 (UTC)

Why not HTML question / relative links
If HTML were used in a wikipedia-like project, wouldn't the links look more like , instead of ? --Biesi 00:10, 8 Jun 2005 (UTC)

Linux distribution
Which Linux distribution does Wikipedia use? I think that would be helpful for the FAQ, too. Massysett 18:20, 17 October 2005 (UTC)


 * Oh, I see--my answer is at Wikimedia servers. Massysett 18:24, 17 October 2005 (UTC)

First server
can i know the history of the 1st server launch into the world? and who create the server — Preceding unsigned comment added by Fujisnow (talk • contribs)
 * See History of Wikipedia.-gadfium 02:22, 21 January 2006 (UTC)

Technical Question about Deletion
If I understand correctly, in wiki software when an article is deleted, the contents remain stored somewhere. Is it stored forever or is it purged after a time and how long until it is purged? Next, if an image deleted is the image data lost for good? I am asking because I'm wondering how wiki software works. DyslexicEditor 07:36, 29 January 2006 (UTC)


 * Deleted edits are still visible to admins, and can be undeleted by an admin (see WP:DRV). Deleted images are gone forever. - jredmond 23:22, 2 March 2006 (UTC)


 * Deleted images are now archived too. Plugwash 12:54, 2 September 2006 (UTC)