Wikipedia:Reference desk/Archives/Computing/2009 February 15

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.

February 15

What computer language(s) is wikipeda coded in

I am assuming PHP/Mysql. but what is the official answer?

142.176.13.22 (talk) 03:37, 15 February 2009 (UTC)

Wikipedia runs on MediaWiki, which is written in PHP. The databases certainly use SQL, I believe you're right that it's MySQL. Algebraist 03:43, 15 February 2009 (UTC)

PHP plus MySQL - and to some extent JavaScript. SteveBaker (talk) 04:04, 15 February 2009 (UTC)

What? Nobody is going to come along and point out that both PHP and MySQL are programs written in another rather common language? -- kainaw™ 04:24, 15 February 2009 (UTC)

And then compiled down to yet another language? – 74  06:00, 15 February 2009 (UTC)

That's irrelevent, incorrect and unnecessarily confusing to our OP. Wikipedia isn't coded in those languages - it's coded in PHP and Javascript. You can look everywhere in the sources and you won't find one single line of C, C++ or machine code. Furthermore, it just happens to be that the PHP interpreter that Wikipedia uses runs on some particular server and that the interpreter was written in some other language - but MediaWiki would run just fine on a PHP interpreter written in (say) Brainfuck and compiled to run on a Beowulf cluster of Babbage analytical engines. Please - don't try to be smart - it doesn't work and it confuses the OP. My first answer was both correct and sufficient. SteveBaker (talk) 18:15, 15 February 2009 (UTC)

If you insist on being technical, Wikipedia isn't coded in any of those; it's written in MediaWiki, as Algebraist helpfully pointed out before your own elaboration (although the line is somewhat blurry). Besides, isn't it just a little hypocritical to be complaining about these responses when you've requested a user rewrite working code that you found ugly, and complained that 'Dvorak needs to "just die already"' in response to a request for the number of Dvorak users? – 74  19:45, 15 February 2009 (UTC)

MediaWiki isn't a "computer language" - it's an application program. The previous responses you link to were not confusing - they weren't actively telling someone something that isn't true. By saying (even indirectly) that Wikipedia is written in C/C++/machine code you are directly telling a confusing untruth. As I said before - if you examine the source code for Wikipedia - you'll find PHP, MySQL and JavaScript - and none of the other things. SteveBaker (talk) 04:21, 16 February 2009 (UTC)

Wikipedia is an encyclopedia, not a program; you wouldn't write an encyclopedia in any programming language. MediaWiki, however, is a program, not an encyclopedia, and unsurprisingly is written in a programming language. So no, Wikipedia (an encylclopedia) is definitely not written in PHP, MySQL, or JavaScript. Anyway, while I certainly don't feel that my comment warranted your rebuke, I'll go ahead and strike it because arguing over the merits of a response to a joke wastes everyone's time. – 74  05:15, 16 February 2009 (UTC)

So in your view, the answer should be "nothing". Great - thanks. SteveBaker (talk) 04:34, 17 February 2009 (UTC)

Ah, another question with a simple answer diluted by people in an endless attempt to be most clever... --98.217.14.211 (talk) 04:32, 16 February 2009 (UTC)

I'm going to stir the pot a little by saying that the only thing a user ever sees is HTML, CSS, and a little JavaScript. PHP and MySQL generate mostly HTML that is served to your browser. So, PHP, JavaScript, SQL, CSS, and C++, but mostly HTML.--K;;m5m k;;m5m (talk) 05:24, 16 February 2009 (UTC)

HTML isn't a programming language - neither is CSS. There isn't one line of C++ code in either Wikipedia or MediaWiki's sources. So the answer is STILL PHP, JavaScript and SQL. SteveBaker (talk) 04:34, 17 February 2009 (UTC)

If you are clever enough to comprehend English, you will see that it was derailed by questioning why nobody has attempted to be clever, not by trying to be clever. -- kainaw™ 13:24, 16 February 2009 (UTC)

Yes - exactly. People too often forget that the object here is to tell the OPs what they need to know. SteveBaker (talk) 04:34, 17 February 2009 (UTC)

Safely Remove Hardware

Does Safely Remove Hardware do anything else besides unmount the filesystem(s)? Thanks in advance. --wj32 ^t/c 03:45, 15 February 2009 (UTC)

I'm not one hundred percent certain, but I'm 99% sure that unmounting the filesystems is all it does. --Aseld ^talk 03:51, 15 February 2009 (UTC)

Well, before unmounting, it flushes the write buffers to disk, but I guess you could also consider that part of unmounting. If nothing has been written to the disk since it was mounted, it should be safe to remove it without using Safely Remove Hardware. Indeterminate (talk) —Preceding undated comment was added at 04:29, 15 February 2009 (UTC).

At least on XP, the "Safely Remove Hardware" menu appears to shutdown and disable USB power to the device; on Vista it apparently does not. (USB is designed to be "hot-pluggable", so disabling power isn't critical. Properly shutting down the device, such as parking disk drive heads, would seem significantly more important.) – 74  05:55, 15 February 2009 (UTC)

Say I have a USB hard disk that requires 2 cables (one for data, one for power) - will unplugging it without using Safely Remove Hardware possibly cause anything more severe than data loss/filesystem corruption? --wj32 ^t/c 09:37, 15 February 2009 (UTC)

Think of all the computers that suffer power failures every day - almost all of those have hard drives, and they survive. While I doubt it's *good* for the drive to cut the power without it spinning down properly, doing it once or twice is not likely to cause any problems. --Aseld ^talk 09:46, 15 February 2009 (UTC)

This may be compounded by moving the drive with the heads improperly parked. On modern drives, however, there is usually a mechanism (a spring, or an electrical reverse feed from the platter motor) to park the heads following a power failure, so no damage would be expected. – 74  20:02, 15 February 2009 (UTC)

On Line Photographs

On some internet pages pictures just have a + on them. How do I view them or what programme do I need to open them219.89.28.66 (talk) 04:25, 15 February 2009 (UTC)

I'm not sure what you're talking about. Could you link to an example? Indeterminate (talk) 04:30, 15 February 2009 (UTC)

I think you may be talking about the auto-resize feature - if you click on the picture - does it then enlarge to full size rather than fit the window? — Ched (talk) 09:04, 15 February 2009 (UTC)

If you're talking about an "X" rather than the "+" (plus) (and no picture) that can be a variety of things - often the picture doesn't exist at the location the web-pages says it does - or you have some sort of blocking going on. If you right-click, and can pick the "View Image", that may help in some cases. — Ched (talk) 09:08, 15 February 2009 (UTC)

Any non-Apple device playback Apple Lossless audio files?

I'm considering ripping my CD collection to consolidate all my music digitally without compression. But are there any non-Apple products that can *use* Apple Lossless songs? I'd like the ability to stream my songs to my PS3 and Xbox. Do they support any lossless compressed audio files? Should I just rip to WAV for maximum compatibility? --71.158.215.154 (talk) 04:39, 15 February 2009 (UTC)

Indubitably. Try VLC or MPlayer. --Aseld ^talk 05:11, 15 February 2009 (UTC)

Oh, unless you mean on the Xbox and PS3 natively without any extra software. --Aseld ^talk 05:12, 15 February 2009 (UTC)

WAV is obviously the simplest option. If space is a concern, you're more likely to have support for Windows Media Lossless on the Xbox360. For the PS3, perhaps FLAC, but seemingly not [1]. Since the PS3 supports BluRay, Dolby TrueHD or DTS-HD Master Audio may be an option although it doesn't seem like there is any free encoder for either yet. Note that the obvious advantage with lossless audio is whether it's WAV, FLAC or whatever, it doesn't matter what you choose, changing your mind 2 years from now shouldn't be too much of a hassle and you're not going to lose quality, so don't worry about it Nil Einne (talk)

Google Translate oddity

Currently, Google Translate seems to translate the Russian word град into the letters "deg", no matter the destination language. Any idea why this might happen? Bart133 ^{t c @ How's my driving?} 05:30, 15 February 2009 (UTC)

Is град a real Russian word? If it's not, it's possible that Translate is just transliterating the Cyrillic. --Aseld ^talk 05:46, 15 February 2009 (UTC)

Check out the disambiguation page. One of the meanings is Grad (angle), maybe google picks deg(ree) because of that. 88.112.63.253 (talk) 06:28, 15 February 2009 (UTC)

The word град has three meanings in Russian-- (1) Grad (angle), (2) Hail, and (3) a city. The last one is a South Slavic word, so it has Old Church Slavonic undertones in modern Russian. As noted, Google is probably mis-mapping град to degree, due to the meaning (1). Google translate is particularly susceptible because it uses something like Statistical machine translation, which is easily misled by homonyms (and even more so by polysemes). Avram (talk) 08:44, 15 February 2009 (UTC)

(This question would get better answers if you asked it on the Language reference desk. SteveBaker (talk) 18:01, 15 February 2009 (UTC))

You tried RusTran? –  LATICS  talk  18:10, 15 February 2009 (UTC)

New Graphics Card

Hey again. I've asked similar questions before, but now I'm sure what I want to ask. I have a HP A6632F desktop, and I want to replace the GPU. It has 4GB of RAM, 2.4Ghz Duo Core processors, and ~452GB HDD. It has an integral PCI-E intel G33/31 Express family chipset already (integrated). Someone said I could probably put a GeForce series 7 or 8 in it, but I was wondering about a series 9. Can anyone recommend a card(s) that would work in my comp? Thanks. --AtTheAbyss (talk) 07:14, 15 February 2009 (UTC)

That model number doesn't seem to exist. What matters most is whether or not you have a PCIe x16 slot. If you do then you can pretty much use any card that doesn't require an external power connector. This includes a number of GeForce 9 series, as well as a number of ATI cards. In the event the PCIe x16 slot doesn't have 16 lanes, you may find performance is limited but I doubt you'd see much difference except with a very fast card (which will require an external power connector, maybe two). If your power supply has a PCIe power connector, then you should be able to use a card which requires one too. If not, then you may be able to use one with an appropriate molex converter but it will depend on the specs of your power supply. If your computer accepts and uses normal ATX power supplies then you can simply upgrade the power supply (make sure it is normal ATX and doesn't use odd wiring or something similar.) Note there are a few other things you need to check. Make sure you have enough room to fit the card. If your case is small some cards may be too long. Also some cards require the slot on the left of the card be free (two slot cards). Finally if your case requires low profile cards, then you'd need a low profile card and bracket. Note if you don't have a PCIe x16 slot, you're pretty much SOL Nil Einne (talk) 07:32, 15 February 2009 (UTC)

BTW, the ATI Radeon 4670 is probably the bestcard which doesn't require an external power connector Nil Einne (talk) 07:53, 15 February 2009 (UTC)

I fixed my model number. --AtTheAbyss (talk) 08:21, 15 February 2009 (UTC)

According to [2] you should have a PCIe x16 slot. From the layour of your motherboard, I don't think your likely to have problems with most cards although there's a slight risk the ram slots may get in the way if it has a very big heat sink which reaches quite far down. Of course, if you want both PCIe 1x slots to be free (or they are already populated) then you'd have to stay clear from a dual slot card. It also looks like your case should accept normal profile cards although I'd verify that myself if I was you. Your power supply however is only 250W [3] and although it should be a decent enough one overall (most branded computer ones are) it's unlikely to have a PCIe power connector and I wouldn't recommend you try using a molex adaptor to get a PCIe power connector since there's a risk you'll be pushing the PSU too hard (it'll probably be fine but personally particularly if you're using an adaptor I wouldn't try). However from my reading your motherboard uses the normal ATX PSU pinout so presuming your case can fit normal ATX PSUs (it looks like it should but I would check more carefully) and the PSU is removable you should be able to upgrade PSU if you do want to use a card which requires an external power connector. Of course you could try a card with an external power connector and see how it goes and upgrade the PSU if you find it necessary (instability etc) P.S. No idea what airflow in your case is like but you may want to look at improving it if you get a card that outputs a lot of heat. P.P.S. [4] may also be useful Nil Einne (talk) 12:24, 15 February 2009 (UTC)

I took some measurements, and my case is 5.75inches wide, and it's 6.75 inches from the rear of the case to the nearest RAM card. I have plenty of room to the side, but my modem does extend 2.75 inches from the rear. I've looked up two cards, and neither fits. Do I need a compact card or something?--AtTheAbyss (talk) 03:21, 16 February 2009 (UTC)

Abbreviated form of terabyte?

When spoken out loud, "kilobyte" is often abbreviated as "K", "megabyte" as "meg" and "gigabyte" as "gig". What's the short form for "terabyte"?121.72.165.197 (talk) 10:19, 15 February 2009 (UTC)

I've always used "T", as in "I want to get one of those 2T hard drives.". --Aseld ^talk 10:26, 15 February 2009 (UTC)

I prefer the slightly longer TB for terabyte and Tb for terabit. I'm not sure what abbreve to use for User:Taraborn, however. :-) StuRat (talk) 17:48, 15 February 2009 (UTC)

The OP was asking for the short form of terabyte for use in speech. Personally, I'd just say "terabyte". When I hear "TB", the first thing that comes to mind is tuberculosis. --173.49.17.152 (talk) 19:51, 15 February 2009 (UTC)

Terabyte-sized things are still pretty rare - so we don't often talk about them and the need to abbreviate never seems to come around - so I just say 'Terabyte'. SteveBaker (talk) 17:59, 15 February 2009 (UTC)

You are behind the times, I believe. One-terabyte external hard drives are rather commonplace these days. --Aseld ^talk 07:50, 16 February 2009 (UTC)

No - I'm not behind the times. 1TB hard drives are certainly becoming more common - but we don't often (if ever) have files that are terabytes long - we don't have terabytes of memory - or data structures that consume a terabyte - or bandwidths that we measure in terabytes per second. So we still don't use the word 'terabyte' very often...not often enough to demand a shorthand form. Really - only in the context of the sizes of hard drives. Also, many of the terabyte hard drives I've seen in Fry's are actually labelled something like '1000 Gbytes' - probably because most people don't know what a terabyte is. SteveBaker (talk) 04:23, 17 February 2009 (UTC)

"Terrs" would follow the form of "gigs", but "terrs" brings to mind for me the line of dialogue in Neuromancer about the "Christ the King terrs", short of "terrorists". Tempshill (talk) 06:04, 16 February 2009 (UTC)

I always refer to them as 'terabyte' when spoken out loud, now when written i refer to them as TB (where Tb would mean terabit)– Elliott(Talk|Cont)  17:16, 16 February 2009 (UTC)

Spreadsheet - automatic date incrementation: 14 days

I would like a column of dates in a spreadsheet file, starting with a specified date (e.g. 2009-02-12) and then continuing with increments of 14 days (next would be 2009-02-26). How can I achieve this (I have access to Excel and Calc). ----Seans Potato Business 13:05, 15 February 2009 (UTC)

In Excel, type the first few dates in order (two should be enough), then select them. A little black square should appear in the bottom right of the lowest selected cell. Mouse over it (your arrow will turn into a black "+"), then click and drag downwards. A tooltop should appear as you drag that tells you what the current cell will be filled with. Let go when the desired date appears. Xenon54 (talk) 13:10, 15 February 2009 (UTC)

Okay thanks; works in Calc too. ----Seans Potato Business 13:45, 15 February 2009 (UTC)

Tables in MS Word 2007

Is it possible to shrink a table of 31 rows so that it fits on one page (in addition to approx. 1/4 of the page above the table being taken up by another smaller table and a few lines of explanation)? The reason I need to do this is because I am working on a translation of a document that is formatted in this way, and have been asked to keep the exact same layout. I have tried to make the font size at its minimum, but it only goes down to 8. Is there anything else I can do?--KageTora (talk) 14:01, 15 February 2009 (UTC)

You can set a smaller font size manually (i.e. type it in the box) but that isn't recommended since 8pt is already difficult to read for some people. You might also be able to change the margins for the table or use a more condensed font to gain a little extra room, and the "AutoFit" context menu might prove useful if you haven't tried that yet. My inclination, however, would be to put the table (by itself) on a landscape page, either the page following the first reference, or in an appendix. If none of those options work, the best bet is probably to change it to a double-line table. – 74  14:48, 15 February 2009 (UTC)

Perl, Regex, Capturing groups

In Perl, capturing groups are numbered $1, $2, $3 ...

The script's own file name is $0.

Other serialized items, such as arrays, are numbered from 0, i.e., $array[0], ..., $array[999].

Isn't it kind of weird? -- Toytoy (talk) 15:05, 15 February 2009 (UTC)

No. Numbering arrays from zero is very common and makes a hell of a lot of sense. The number in an array is an offset from the first element of the array. So, the first element is offset by zero. The next is offset by one. Regex groups are not numbered as an offset. They are matches, first match, second match, and so on. So, it makes sense to number them from one. I'm sure someone will argue that it makes sense to number arrays from one and then give examples of languages that do so, but that doesn't mean that numbering arrays from zero is senseless. It just means that someone else did it different. -- kainaw™ 15:39, 15 February 2009 (UTC)

If you think it makes sense to number arrays from 1, Perl's got you covered: set $[ to change the index of the first element in an array and the first character in a substring. If you prefer to start at 7, Perl will presumably accommodate. How many other languages allow you to index from an arbitrary base value? Note: modifying $[ is highly discouraged. – 74  16:25, 15 February 2009 (UTC)

Fortran. - SigmaEpsilon → ΣΕ 19:32, 15 February 2009 (UTC)

I should've guessed.   ;-)   – 74  19:51, 15 February 2009 (UTC)

There are many - even popular ones, such as PHP. If you declare $myarray[7] = "Hi"; Then, using something like $myarray[] = "World"; will set index 8 (one more than the highest number already set). What does this really mean? Languages like PHP and Perl are not using arrays as defined years and years ago. They are using hash tables with incrementing numeric indexes, but loosely referring to them as arrays. From the programmer's point of view, it doesn't matter. From the computer's point of view, a hash table and an array are two very different things. -- kainaw™ 20:18, 15 February 2009 (UTC)

While I agree that the result may be very similar (or even identical), there is a difference between creating an array whose meaningful values start at "7" and base-indexing all arrays from 7. By the former requirement, any array with the first 7 values remaining unused would qualify (which practically includes every language with array support), while the latter requires all arrays (including the return from functions like 'split') to be indexed from 7 by default. In many respects a per-array implementation is more useful, but allowing modification to the underlying base index value is (to my knowledge) significantly less common, probably because of its questionable value and tendency to confuse. – 74  22:21, 15 February 2009 (UTC)

Capturing groups in regexes in every other languages start from 1 (perhaps following from Perl). Group 0 by convention stores the entire match in languages other than Perl (weirdly, in Perl this is called $&). Examples: Python, Ruby, JavaScript, Java, PHP, Haskell. --Spoon! (talk) 20:15, 15 February 2009 (UTC)

A glance at our article on AWK - one of the predecessors to Perl - suggests that this special use of "zeroeth group meaning whole match" is closest to the origin of this design decision in Perl. In AWK, $1, $2, etc, refer to fields of the input line, but $0 refers to the whole line. Since "capture groups" in a Perl regular expression are broadly equivalent to AWK's "fields", especially for simple, non-nested groups, it would have been confusing if $1 referred to the second such group. - IMSoP (talk) 23:50, 15 February 2009 (UTC)

I am still new to Perl programming. Is there a simple way to retrieve the whole line containing the match. Let's say I want to show all lines (entire lines) containing the word "dog" in a long article, how do I do it? I know I can use \n to mark the boundaries of a full line. With a little extra programming, I can get the sentence containing the word. I just want to know is there a magical way to do it. Because regex expressions are difficult to read and debug. I always want to keep them as short and simple as possible. -- Toytoy (talk) 00:50, 16 February 2009 (UTC)

Normally you're working a line at a time anyway, so it would look like:

while (<>) {
  print if /dog/;
}

--Sean 14:29, 16 February 2009 (UTC)

AM2+ cpu

Hi, how new is AM2+ compared to AM2? Is it only now released? Was it available last summer? ~ R.T.G 15:37, 15 February 2009 (UTC)

AM2+ is an upgraded AM2. I don't know when the very first AM2+ motherboards hit the market, but there were plenty available last summer. -- kainaw™ 15:44, 15 February 2009 (UTC)

Thanks Kainaw. Anyone with some good reference material might want to give a bit of attention to AM2+ as it is very short and barely 50 edits since over 2 years ~ R.T.G 17:46, 15 February 2009 (UTC)

AM2+ mobos will let you use AM3 CPUs. That's pretty much the most useful part. 24.76.160.236 (talk) 10:30, 16 February 2009 (UTC)

Web hosting

I'm looking for a cheap web host (around USD 60 per year or less — though free is good too :-D) that provides the following:

• At least 200 MB of server space
• A decent amount of bandwidth (however much that is)
• The ability to upload already-coded HTML files and maintain an existing folder structure
  • For example, I have my HTML files in a folder called 'web', and the code points to the path 'images/thumbnails/filename.jpg' — I want to be able to maintain that and not have to re-code it
• CSS and JavaScript functionality
• The ability to use a picture for the background (specified in a separate CSS file)
• The ability to use a domain name that I already own, not a subdomain

Does anyone have any suggestions? Thanks, Hermione1980 16:51, 15 February 2009 (UTC)

Have you looked at godaddy.com? — Ched (talk) 17:40, 15 February 2009 (UTC)

Any worthwhile hosting package will do most of those (only the first two vary); only toy things, which are gnerally free, won't. Dreamhost will do everything you specify, for roughly your target price. That said, you'll find people who are unhappy with Dreamhost, and indeed with godaddy. $60USD is a very small amount of money, so I'm pretty much impressed with what you do get for that. 87.112.81.29 (talk) 20:23, 15 February 2009 (UTC)

Bluehost is also a pretty popular cheap web host. And yeah, you'll find people who are unhappy with everything, but personally I've never had a problem with Bluehost, and I've been using it for years for manage tons of small sites. --98.217.14.211 (talk) 20:31, 15 February 2009 (UTC)

Generally speaking, the cheap ones will have slower hosting (especially slower SQL, and slower scripting). Otherwise, they'll work fine. The things you've asked for (CSS and JS?) are pretty such impossible for a web host not to offer, considering they're all client side... 24.76.160.236 (talk) 10:28, 16 February 2009 (UTC)

You should read the comments on web hosing review sites and see they have been in business for a few years. I've everything you say plus multiple domains SQL and python and a few other goodies for less and the host is fast and very reliable and response time for problems is very good, so you're not asking for too much. The one I used till a couple of years ago though I just got the cheapest and it was not worth even that small amount. And yes avoid the free sites. Dmcq (talk) 13:46, 16 February 2009 (UTC)

Heh. Really the only reason I specified the CSS/JS was because I started with Google Apps...did not realize until after I signed up that they have virtually zero features for actual web design. :-( I'll definitely be reading the reviews. Thanks, guys. Hermione1980 15:29, 16 February 2009 (UTC)

Just an idea, but if you have access to an internet connection thats always on, then you maybe able to host the webpage your self. Even if you dont have a static IP address... Just Throwing it out there. – Elliott(Talk|Cont)  17:13, 16 February 2009 (UTC)

Space between lines on Word 2007

I have noticed that between each line on Word 2007, there seems to be enough space for another line. Is this the default? I never noticed it on Word XP or 2003. I am wondering if my settings have been changed, as I made a table with eight rows on it, then merged the rows. After doing this, I could only fit 4 lines of text in. Is there any way to change this?--KageTora (talk) 18:44, 15 February 2009 (UTC)

This is called "line spacing" and it's fully configurable. (It's located under "line spacing" on the "Paragraph" context menu on an older version of Word.) – 74  19:55, 15 February 2009 (UTC)

Similarly, hitting return ends a paragraph (a hard return), whereas in older versions Microsoft Word) it would add a line break (a soft return). Cycle~ (talk) 21:44, 15 February 2009 (UTC)

I'm 99% sure that's not true. The return key always added a paragraph break, denoted by ¶ (a pilcrow) if you enabled "show hidden characters"; a soft return could be entered by holding shift, and would be denoted by ↵ (the symbol on most carriage return keys). I imagine this is still the case.

The difference, I imagine, is in the default paragraph styles - since most versions of MS Word default to not adding extra vertical space above and below paragraphs, users commonly hit return twice to artificially produce vertical space between paragraphs, in fact creating an empty paragraph in between.

Incidentally, the awkwardness of changing this and other style behaviours in MS Word is one of the reasons why I have come to actively prefer the interface of OpenOffice.org, which has a floating tool palette for selecting and editing the named styles. (Exactly the same is possible in MS Word, but the interface is more cumbersome.) - IMSoP (talk) 23:42, 15 February 2009 (UTC)

Yes, that is the default style. If you look at the ribbon of buttons, one set is marked "Styles" and the first (ie. default) style is "Normal" which is defined to include 1.15 line spacing and a 10pt gap after each paragraph. To change the style of the current paragraph, click the "No spacing" style which is on the ribbon next to the "Normal" button. To change it permanently, you will need to modify the "Normal" style and save it in the "default.dotx" template. Astronaut (talk) 02:14, 16 February 2009 (UTC)

A simpler way to get back those settings that many of us are used to is to look for the Change Styles option on the Home ribbon, click on it, select Style Set, and then choose Word 2003. -- Tcncv (talk) 06:42, 16 February 2009 (UTC)

How to extract URLs from a Wikipedia article?

I would like to get a list of URLs that are being used as references within a Wikipedia article. Since references are inline, manually going through the text would take some time. Wikipedia is very popular so I'm wondering if anyone has already made some sort of utility or script that does the parsing? A Quest For Knowledge (talk) 20:43, 15 February 2009 (UTC)

All properly formatted articles have a "references" section near the bottom that places all the references in a very easy to cut-n-paste section. If you are looking at an article that does not have a references section, please let us know so we can fix it. -- kainaw™ 21:57, 15 February 2009 (UTC)

Opera provides a table of links on the current page, which might be what you're looking for. It seems like this functionality would be a relatively simple Wikipedia special page, but we don't appear to have anything equivalent. (The References section, if properly present, provides a list of links, but not necessarily a list of URLs.) – 74  22:18, 15 February 2009 (UTC)

The article I'm interested in is this. Yes, there's a references section but most of the references displayed on the page show the article's name, not the URL. I'll give Opera a try. A Quest For Knowledge (talk) 22:21, 15 February 2009 (UTC)

You're talking about the Tools | Links feature, right? Hmmm...it gives me a list, but it looks like Opera is using a list view control and I need text. There doesn't seem to be a way to save this list as a text file. I don't know anything about Wikipedia special pages but I can write a C# app that parses the text. A Quest For Knowledge (talk) 22:33, 15 February 2009 (UTC)

Once you get the list page, Select All and Ctrl+C, otherwise Select All, Right click and choose "Copy link address". Then paste the contents to a text file. I got 890 links for the article you mentioned. I'm using Opera 8. Jay (talk) 08:50, 16 February 2009 (UTC)

Alright, here's a simple javascript function that collects links and throws 'em at the end of the page:

function link_find() {
	var linkss = "";

	for( var i = 0, j = document.getElementsByTagName('a'); j[i]; i++ ) {

		if (j[i].href && !j[i].href.match(/en.wikipedia.org/) && !j[i].href.match(/javascript/)) {
			linkss = linkss + '<br>' + j[i].href;
		}
	}

	var mydiv = document.createElement('div');
	mydiv.style.position = 'relative';
	mydiv.style.top = '0px';
	mydiv.style.right = '0px';
	mydiv.style.border = '1px solid #000';
	mydiv.style.backgroundColor = '#fff';
	mydiv.style.color = '#000';
	mydiv.innerHTML = linkss;
	document.body.appendChild(mydiv);
}


document.addEventListener('load',function addButton() {

	// throw a link up in the corner

	if( !document.body ) { return; }
	var mydiv = document.createElement('div');
	mydiv.style.position = 'fixed';
	mydiv.style.top = '0px';
	mydiv.style.right = '0px';
	mydiv.style.border = '1px solid #000';
	mydiv.style.backgroundColor = '#fff';
	mydiv.style.color = '#000';
	mydiv.innerHTML = '<a href="javascript:link_find();">links</a>';
	document.body.appendChild(mydiv);

},false);

This code is intended (and tested) to run in Opera, but I'm sure it could be adapted to greasemonkey with a minimum of effort. You can modify the regex checks to exclude any certain URLs you desire. – 74  00:01, 16 February 2009 (UTC)

Thanks to everyone for their help! A Quest For Knowledge (talk) 14:28, 16 February 2009 (UTC)