Wikipedia:Reference desk/Archives/Computing/2009 December 22

= December 22 =

IRC Channels in Pidgin
How can I connect to #wikipedia-en-help using pidgin? I can't seem to figure it out. —Preceding unsigned comment added by MMS2013 (talk • contribs) 02:04, 22 December 2009 (UTC)


 * First, you need to connect to an IRC service. I suggest freenode.  Make a new account in Pidgin:
 * Protocol: IRC
 * Username: Whatever you like
 * Server: irc.freenode.net
 * Password: blank
 * Local alias: blank
 * Port: 6667
 * Encodings: UTF-8
 * Everything else, leave as default - meaning don't change it.
 * Now, connect to the new account. Go to "Join a Chat".  Select the freenode account. Enter "#wikipedia-en-help" in the channel box.  Leave the password blank.  Click "join".  I know that sounds like a LOT of work, but it isn't.  It is two steps.  Connect to an IRC service.  Join the chat. --  k a i n a w &trade; 02:09, 22 December 2009 (UTC)
 * "Suggesting" freenode isn't an option, it's mandatory - #wikipedia-en-help is not populated on other IRC servers. Coreycubed (talk) 16:18, 22 December 2009 (UTC)

script help
I had the following greasemonkey script that added the page name to the title bar, for example if the url was " http://example.com/page/123.html " the page title would be "example.com - 123". This is the script: if (location.href.match(/([^/]+)\.html$/)) document.title = 'example.com -' + RegExp.lastParen; However, the website seemed to have changed one tiny thing; urls no longer have ".html" at the end, so now it's just " http://example.com/page/123 " and my script doesn't work. Please could someone have a look at it and tell me what I need to do to get it working again. Thank you! —Preceding unsigned comment added by 82.43.88.54 (talk) 09:56, 22 December 2009 (UTC)


 * Well, I'm not sure exactly what you are going for, but it should be trivial to simply optionally strip the .html (or .htm) extension:

if (location.href.match(/([^/]+)(\.html?)?$/)) document.title = 'example.com -' + RegExp.lastParen;
 * Is that what you are looking for? &mdash;ShadowRanger (talk 15:12, 22 December 2009 (UTC)
 * Thank you, but it didn't work :( —Preceding unsigned comment added by 82.43.88.54 (talk) 15:25, 22 December 2009 (UTC)
 * Oops. Just realized you used LastParen. That would copy out the .html, if it existed. I don't know JavaScript all that well, but you should change the second line to pull the first capture group. &mdash;ShadowRanger (talk 15:52, 22 December 2009 (UTC)
 * Could you explain a little more, first capture group?
 * Each parenthetical expression is a capture group in a regular expression like the one you are using. Because I put parentheses around the .html, I created a second capture group, so the RegExp.lastParen is pulling in the contents of that group, not the one you want. If JavaScript supports Perl-style RegExs (I don't know offhand if it does), you could fix it simply by adding a "?:" in the html capture group, like so:

if (location.href.match(/([^/]+)(?:\.html?)?$/)) document.title = 'example.com -' + RegExp.lastParen;
 * If it doesn't, you'll need to change RegExp.lastParen to a different getter that gets either the first capture group, or explicitly allows you to select by numerical index. &mdash;ShadowRanger (talk 16:45, 22 December 2009 (UTC)

YES!!!!!!! THANK YOU!!!!

Got myself into a mess with Domains and Workgroups
My Windows XP PC was on a the domain for my company. I was experimenting with workgroups (so thatI could share a printer) and moved it onto a local workgroup. However, it is now booting into what seems to be the default account for the PC (an IT admin who is no longer with the company). When I try to re-attach it to the domain, it won't do it because it can't contact the domain controller (the PC is not in the office). Now, I am sure that I can eventually figure out how to connect over the VPN and re-connect to the domain controller, but is there a way to go back on the domain without having to contact the domain server? If not, is there way to have the PC boot into my original account (so that I get the same desktop etc) while staying in the workgroup. Any help would be really appreciated, because getting hold of the VPN client at the moment could take a few days (due to company restructuring and Xmas holidays). LooseJuice (talk) 10:29, 22 December 2009 (UTC)


 * If you just want access to the files on your other desktop, on XP they're usually located in C:\Documents and Settings\USERNAME\Desktop but to rejoin the domain is really going to require access to a domain controller. I don't know if this will actually work (it's a pretty almighty kludge if it does!), but you could try using System Restore to roll your computer back to a date where you were still on the domain. Since you're not actually actively talking to the Domain Controller this may work enough for you to be able to log into your normal account and use it as before, but you may have problems later on when you try and talk to the domain resources/computers. If you try this, please do let us know if this works as I'd love to know! Regarding the later possible problems, when you leave a domain your computer tells the controller that it's leaving and the computer account is then automatically removed from Active Directory and if that's happened then even if System Restore works then your machine won't be able to authenticate with some network resources until it's rejoined to the domain, however since you're unable to connect to the domain controller now I'm assuming it wasn't able to connect at the point you dejoined it from the domain (you get an error message saying about computer account data couldn't be removed from the domain) so this may not affect you. Hope this is of some help! ZX81  talk  15:52, 22 December 2009 (UTC)


 * You may also be able to create a local user account and copy the domain account folder's contents to the new local account. You may need to log in first as the new user to get Windows to initialize the user folder for you. Then, reboot and log in as your first local account and do the copy. That way no files will be locked in the process. Coreycubed (talk) 16:13, 22 December 2009 (UTC)


 * I've found that XP is so horny to restore the default workgroup settings that I just gave up and let it use the default, and put everything under that. This may be fixed in Service Pack 2 or 3, though, as they seem to have made improvements in this area.  StuRat (talk) 23:34, 22 December 2009 (UTC)

Internet browser speeds
As part of it's advertising, Google sometimes refers to it's Chrome browser as 'The world's fastest browser', or so it was said on a billboard I travelled past. My question is, how much of your overall internet speed depends on your browser and how much depends on your connection? As a result, what would be the fastest free browser I would be able to get? Chevymontecarlo (talk) 11:04, 22 December 2009 (UTC)

I happened to have timed six browsers on my computer a few weeks ago to see which started up the fastest. The following are in seconds, followed by milliseconds:


 * 1) Internet Explorer 6: 800 ms
 * 2) Internet Explorer 8: 2400 ms
 * 3) Opera 9: 2700 ms
 * 4) Konqueror: 3100 ms
 * 5) Safari 4: 3200 ms
 * 6) Firefox 3.5: 4200 ms

Another benchmark is how long it takes a browser to complete the Sunspider benchmarks. Those measure the speed with which a browser processes JavaScripts:


 * 1) Safari 4: 615.0 ms
 * 2) Firefox 3.5: 1054.4 ms
 * 3) Opera 9: 4053.8 ms
 * 4) Konqueror: 4756.0 ms
 * 5) Internet Explorer 8: 5536.8 ms
 * 6) Internet Explorer 6: 60788.8 ms

I'm not sure about HTML and image-loading speeds, though.--Drknkn (talk) 12:05, 22 December 2009 (UTC)

Thanks for sharing that research. Perhaps I'll Google around a bit more and see what I can find. Chevymontecarlo (talk) 12:11, 22 December 2009 (UTC)


 * Internet Explorer startup speeds are known to be false. Much of Internet Explorer is shared with Windows.  So, just starting Windows starts up half of Internet Explorer.  The other browsers need to start from scratch.  Firefox has some kind of jump-start thing that will start most of it when the computer starts so you can launch Firefox quicker. --  k a i n a w &trade; 12:20, 22 December 2009 (UTC)


 * The IE speeds are hardly false - they are just able to access fringe-benefits by being part of the main operating system. It seems like that's worth noting as a caveat for speeds, but it doesn't make their speed invalid in anyway (the speeds are important in so much as they are about the end-user's experience aren't they?). 194.221.133.226 (talk) 13:50, 22 December 2009 (UTC)


 * If you include the extra time it takes for the necessary components of IE to start when Windows starts, you have an accurate start time. The user does have to wait for Windows to start up.  That startup time takes longer because half of IE is started when Windows starts.  So, artificially deciding to start all timings after Windows starts provides a false/invalid value. --  k a i n a w &trade; 13:57, 22 December 2009 (UTC)


 * I don't see how that's relevant though? You have to wait for Windows to start up before you can use *any* application/browser, so by your definition you should be applying that startup time to all browsers? After all the other browsers will also need some of the components that Windows loads whilst starting up... ZX81  talk  15:38, 22 December 2009 (UTC)


 * I don't know how to make it more understandable. Windows startup time includes time spent doing nothing else except getting IE ready to start.  If IE didn't exist, Windows startup time would be faster.  So, that extra time spent getting IE ready is technically part of the startup time for IE.  Just because you are forced to do it when you start windows doesn't mean it magically doesn't happen.  Your argument is like claiming that if I spend 12 hours walking most of a marathon the day before the marathon, leaving me only a few inches to run, I can get an official runtime of a second for taking that last step when they fire off the starting pistol.  Any time spent working on something counts as time spent working on it.  For IE startup, it doesn't matter if it takes place while Windows is booting or after the user clicks the shiny blue E.  Time spent getting IE started is time spent getting IE started. --  k a i n a w &trade; 16:34, 22 December 2009 (UTC)


 * It's my understanding that Windows doesn't load parts of IE when it starts up, but rather IE uses parts of Windows that have already started up (for other reasons). For example Trident, IE's layout engine and probably the largest part of the browser, is used by Windows itself whenever it needs to display a HTML Application as well as any other applications that use it. ZX81  talk  16:56, 22 December 2009 (UTC)


 * That's a good trick. Instead of calling the HTML layout engine part of IE that Windows uses, call it part of Windows that IE uses.  This all goes back to that old anti-trust case in which Microsoft claimed you could remove IE from Windows, but it turned out that Windows was using so much of IE that it couldn't be removed. --  k a i n a w &trade; 17:22, 22 December 2009 (UTC)


 * Kainaw, surely you understand that to any user on Earth other than browser authors, the relevant time is from double-clicking the icon to the browser window being navigable. Comet Tuttle (talk) 17:29, 22 December 2009 (UTC)
 * Windows takes the same amount of time to load whatever browser you are going to use on it once it finishes, so to the end user it makes no difference. You still have to wait while the IE components load regardless of whether or not you intend to use IE. It's a good reason to complain about Microsoft, but it isn't a good reason to reject IE. --Tango (talk) 18:02, 22 December 2009 (UTC)


 * I wasn't rejecting IE. I was stating that the startup time is not valid.  Parts of IE (or the parts of Windows that IE uses) are already started when the user clicks the icon.  Starting up those parts of Windows that IE uses makes Windows take a little longer to boot up.  That extra boot time is, in my opinion, IE startup time.  That is all. --  k a i n a w &trade; 21:38, 22 December 2009 (UTC)


 * I contend you need to step back about 10 paces and reconsider your use of the term "valid". Valid in what way?  It's certainly "valid" for the consumer who just wants a browser that launches quickly.  The only way it's not "valid" is if you are a browser developer whining about how it's not fair that your browser launches twice as slowly as IE.  Comet Tuttle (talk) 23:48, 22 December 2009 (UTC)


 * By that argument - just add whatever browser you like to the startup folder so it launches automatically when you start Windows. Then, the startup time, once Windows finishes, will be zero seconds.  Everyone wins. --  k a i n a w &trade; 02:10, 23 December 2009 (UTC)
 * But then your computer will take longer to load. Having Windows load IE components at start up doesn't make Windows take longer to load in a meaningful way because there is no alternative to compare it to. You should only compare alternatives that actually exist. Two alternatives that do exist are "loading IE at startup and using IE" and "loading Firefox and IE at startup and using Firefox". In both cases your browser loads quickly, but in the latter case your computer takes longer to boot. "Loading only FF at startup and using FF" isn't something you can do so it shouldn't factor into your comparisons. --Tango (talk) 16:00, 23 December 2009 (UTC)


 * You are now claiming that IE partially loads at startup, which is what I said at the very beginning. So, it appears that you agree.  IE partially loads at start time.  What's the argument? --  k a i n a w &trade; 22:43, 23 December 2009 (UTC)


 * The argument is (and has been from the beginning) about whether that's relevant in the slightest. No one debates that certain IE componants are loaded with windows. The point is that any measure of browser start-up time that's even remotely useful to anybody at all, will measure the time from clicking on the icon to the time when you can use the browser.
 * Users (and hence, the question-asker) don't care if IE has a head start. But if it crosses the finish line first, that's a point in its favor. A minor one, though, because page-load is more important than browser-load. APL (talk) 01:38, 24 December 2009 (UTC)


 * Anyway, I'm not sure that startup time is really all that important. As the numbers above indicate quite clearly, just because something starts up quickly does NOT by any means indicate it will be fast at anything else. IE6 starts up mighty quick, but is a total slouch when it comes to rendering and Javascript and etc. So I do think that much of the above argument is rather moot from a user perspective! --Mr.98 (talk) 17:34, 22 December 2009 (UTC)


 * I can think of one case where it matters; if you view lots of HTML docs, and it starts up a browser each time you pick one to view (then you close the browser and click on the next doc, which opens up the browser again, etc.). However, most people probably do something more like the tasks measured by the 2nd set of benchmarks, which show IE to be up to 100 times slower.  That's huge.  As for whether IE's quick start-up time "counts", I'd say it counts when choosing a default browser for viewing docs, but not when IE claims their browser is better.  I also find myself wondering about differences in the initial start-up time and repeat start-up times in the same session.  That is, do the browsers leave something in memory to make subsequent start-ups quicker ?  The downside, of course, would be filling up the computer's memory (behaving much like a memory leak, even though it may be intentional, in this case). StuRat (talk) 23:22, 22 December 2009 (UTC)


 * The problem is that the benchmarks have to a bit more varied than what we've got on here. There are lots of blogs that discuss this. IE6 is 100X slower if what you are doing is string manipulation in JS. IE6 is especially bad at that (and it is not an uncommon thing to do with a web page, so that is a real problem), but it is not as bad with a lot of other things. Similarly, depending on what you are measuring, even just within the general area of JS activities, you get very different results. (Of course, IE6 is now two versions out of date, and has been notoriously buggy in all respects for ages, so I do think that it's a bit of a straw man, as nobody who really cares about browser performance should still be using it, but anyway.)
 * Note that some software has the possibility of a preloader if you need fast startup time—Google "Firefox preloader" and you can find plugins that do the exact same thing. --Mr.98 (talk) 15:50, 23 December 2009 (UTC)

failure to recognize DVD disc
My DVD player sometimes fails to recognize a DVD disc, though if I try again, maybe several times, it will load properly. This also just happened on a computer just now with system restore discs. What is happening? How does the DVD initialization process work? Is there something I could do here other than retrying? --Halcatalyst (talk) 20:56, 22 December 2009 (UTC)


 * Hmmmm....You're question is a little unclear. Maybe you should try buying a dvd drive cleaner disk.  What they do is clean the lens and tests the drive for functionality.  Most disks usually have a small brush on them.  Btilm Happy Holidays! 21:08, 22 December 2009 (UTC)
 * Thanks for the hint about the DVD cleaner disk. My question arises mainly out of curiosity: How does the DVD initialization process work? I would like to have some kind of idea what is going on in there when it fails. I assume there is a problem reading a particular sector. What might be the nature of the problem and why would it eventually go away? --Halcatalyst (talk) 23:11, 22 December 2009 (UTC)


 * Warning: I always thought those DVD cleaner discs and CD cleaner discs were a scam that damaged your equipment.  Comet Tuttle (talk) 23:46, 22 December 2009 (UTC)


 * There is probably some risk, I wouldn't expect the bristles themselves to be that likely to scratch the lens unless the designer really screwed up but the dust on the lens may. Opening the drive and cleaning manual is obviously better but not always feasible. The laser lens can definitely get dirty in some cases so cleaning in some way may help if that's the problem. However anything which claims you need to regularly clean the lens or makes it sound like you're going to get better quality audio or video from cleaning is mostly bullshit Nil Einne (talk) 00:54, 23 December 2009 (UTC)


 * One problem I've seen is if the DVD isn't quite in proper position. This is more common on mini-CDs, since they have to stay in that tiny depression in the center, but can also happen with full sized CDs/DVDs, occasionally.  I like to wiggle them around after I set them in the tray, which helps me to tell if they are seated properly (if not, they slide more). StuRat (talk) 23:04, 22 December 2009 (UTC)

The search terms in the Google URL
The long URL that comes with a page of Google search results seems to be split up into parts that consist of "&", some letters, a "=", another text-string. Where can I find information about what these various parts do? I am particularly interested in Image search - in particular I'm wondering if its possible to get Google to do a related image search while still being able to specify the image size. Thanks. 84.13.44.90 (talk) 21:24, 22 December 2009 (UTC)


 * Well, the key one that controls size is imgsz (e.g. imgsz=l is large). But plunking that on to the end of a "Similar" search does not work. My guess is you can't combine the two methods of searching at all. Whether that is just because Google doesn't let you, or because of some way they store the image data, I don't know. --Mr.98 (talk) 21:41, 22 December 2009 (UTC)


 * On man's collection of google search parameters, fwiw. --Tagishsimon (talk) 23:10, 22 December 2009 (UTC)

Filenames not displayed in Filmstrip view
I use Windows XP Pro. On my PC there is one folder of images for which Windows Explorer does not display filenames in Filmstrip (or Thumbnail) view. To troubleshoot the problem I copied the images to a new folder; then filenames were displayed correctly in the new folder. But when I deleted the original folder and renamed the new folder to the original name, the problem recurred. So apparently Windows associates the problem with the folder name. Does anyone know how to fix this? (I do want to keep the original folder name.) 76.169.63.170 (talk) 22:27, 22 December 2009 (UTC)

restoring Vista backup to XP
Having a Vista computer with all kinds of problems, I decided to do a system recovery. The original OS was XP, so that of course is what I have now. I have the discs to convert again to Vista, but I was not thrilled with it and would like to remain with XP if I can.

Trouble is, I need to restore user data I backed up using Vista Backup. The XP recover utility doesn't seem to recognize that backup. Is there any way to recover the data other without reinstalling Vista again? --Halcatalyst (talk) 23:24, 22 December 2009 (UTC)


 * I initially thought there was just no way this would work, but this link discusses how to "rewind" from Vista back to XP on the same machine ... so it may be possible for you to do your system recovery under Vista and then attempt to rewind to XP. Personally I would probably get a new hard disk and start from scratch, changing your current hard disk from the C: drive to the D: drive, and moving its documents over to the C: drive after the fresh install.  If you've got money to burn on a new hard disk.  Comet Tuttle (talk) 23:40, 22 December 2009 (UTC)


 * I believe - based on limited experience with Vista Home Basic - that the backup files are mostly zip files, which can be read using any appropriate software (including XP Windows Explorer's built-in handling). Finding individual files might be tedious, but the data should be recoverable. Mitch Ames (talk) 00:08, 23 December 2009 (UTC)