Wikipedia:Reference desk/Archives/Computing/2010 June 24

Computing desk
< June 23 << May | June | Jul >> June 25 >
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.


June 24

edit

Operating System Advice

edit

I'm planning on building my first computer soon, but I'm still not sure what OS to use. I was hoping someone here would have more experience with OSes and be able to offer a little advice to a poor lost soul such as myself. Basically, I'm deciding between Windows 7 and Windows XP. (please don't tell me I should get a Mac or Linux; I really don't want a Mac and I might dual-boot with Linux anyway, so you don't need to proselytize) I plan to use the computer mostly for everyday work (word processor (which I can use Open Office for, to save on buying Office), the Internet, that sort of thing), but also playing the occasional game, mostly Valve Corporation stuff. I don't really know anything about operating systems, though. So which would be better for playing games, and which version of the operating system should I get? I've heard Windows 7 has some trouble playing games, especially older ones, but are they enough of a concern for me to go with XP? Or are there other factors that might make me want to choose one over the other?

As a side note, while money isn't too big of a concern here, it's still a consideration. I'm a student, so I think I'd get a discount on Windows 7, but I'm not sure about XP. So... any help? Thanks. Fletch the Mighty (talk) 03:44, 24 June 2010 (UTC)[reply]

I upgraded a little while ago from XP to windows 7 and have never looked back, and I really liked XP. This is obviously just my opinion. I work in IT so have considerable exposure to computers and operating systems. One main advantage of W7 is Direct X 11, however that's only with recent games, not old ones, and I can't say I play a lot of old games, so that part I don't really have an opinion on, I've found compatibility mode to be pretty good. The few things I haven't been able to get working on W7, like some pretty specialized programming hardware and my serial to USB dongle I've worked around by installing XP on a virtual machine which I run using the free Vmware player. However you still need a license of XP to legally install it on a virtual machine, and I haven't tried playing games on it, but if the games are old and your computer is powerful, i think it would be alright. Vespine (talk) 03:59, 24 June 2010 (UTC)[reply]
Keep in mind that Microsoft is phasing out support for Windows XP, and it hasn't been available to purchase for about one to two years now (except for "ultra low-cost PCs", where it's available until October). While security fixes will still be provided until early 2014, technical support and feature upgrades have been discontinued for over a year, unless you have already purchased an extended support contract. - Bottom line is that if you don't have an XP license already, you're unlikely to find one. Even if you do, for long term support, you're better off going with Windows 7, unless your hardware doesn't support the Win7 system requirements. -- 174.24.195.56 (talk) 04:26, 24 June 2010 (UTC)[reply]
Windows 7, hands down. Simply the "Search" field off the Start menu is great. It crashes far less due to the different driver model. Comet Tuttle (talk) 05:41, 24 June 2010 (UTC)[reply]
Adding my support for Windows 7 too. Over the last ~6 years I've used Valve's Steam platform for pretty much all of my games (just counted - 132 different games are in my list!) and I haven't had any major problems (although I didn't have any problems with Windows Vista either for that matter!). There are a handful of the older games that have a few niggles/need slight tweaking to get to work properly (you can get the relevant information from the Steam forums), but anything released in the last 2-3 years should run problem-free, even on Windows 7 x64 which I'm using. Hope this helps! ZX81 talk 12:14, 24 June 2010 (UTC)[reply]
Anyway avoid Vista. I had a notebook with it which died irreparably recently and I was so glad because it meant I could go and get a new notebook with WIndows 7 instead which I'm quite happy with. Dmcq (talk) 13:45, 24 June 2010 (UTC)[reply]

OK, thanks for the advice, everybody! I'm definitely going to go with Windows 7. I didn't think about the support side of it before, but 174.24 above is right, so thanks for that! Fletch the Mighty (talk) 23:11, 24 June 2010 (UTC)[reply]

Networking

edit

How to control a PC with hardwares of some other PC connected with it on LAN? 220.225.96.217 (talk) 04:30, 24 June 2010 (UTC)[reply]

If you are looking for software to remotely control one PC from another, VNC software such as TightVNC (preferably with the "mirror display driver") or UltraVNC should do the job. Alternatively, Windows has a built-in Windows Remote Assistance feature that can be used for remote access (the main limitation is that someone has to be at the PC being controlled to accept the connection). Finally, some versions of Windows have "remote access" or "Remote Desktop" settings in the System control panel that will allow you to connect from the other computer using Remote Desktop Connection (the main limitation is that the connection will not work if you are logged in at the time). PleaseStand (talk) 05:26, 24 June 2010 (UTC)[reply]
Actually if you're using version 5 of Remote Desktop Connection client there's a /console parameter to connect to the main console and in version 6 this changed to /admin. Running the client with this parameter (i.e. mstsc /admin ) with connect to the main console of the target machine. However, only if you have the same credentials as that user that's already logged on. ZX81 talk 11:55, 24 June 2010 (UTC)[reply]
Also, fwiw, Windows XP, Vista, 2003, 2008, and Windows 7 all have Remote Desktop capability. Some simply restrict the number of users that can connect at any given time, which often forces you to take control over the "main desktop" session. The setting for this is available if you right click "My Computer" (just "Computer" in 2008/Win7) and select "Properties", there will be a Remote tab (or "remote settings" area in 2008/Win7) that lets you activate or deactivate the feature. Nothing personal to anyone, but it astounds me that so many people use VNC when this built-in, tried and true method of remotely accessing a windows computer is right there ready to go. --144.191.148.3 (talk) 13:55, 24 June 2010 (UTC)[reply]
Windows Vista starter, home basic, and home premium editions do not feature an RDP server, and neither do their counterparts in Windows 7. -- Finlay McWalterTalk 16:54, 24 June 2010 (UTC)[reply]

how to boot from cd?

edit

Today I got my newly built computer to power up. It asked me to insert a cd. So I inserted a Linux cd I downloaded from the internet. (I need to wait until my friend gives me back my Windows cd so I figured I'd try to use Linux instead.) But it was a bad file I downloaded and it didn't install good. It froze and the screen was black. So after a while I restarted the computer but it isn't the same as before. It doesn't ask me to insert a disk anymore. Instead it says verifying dmi pool data and just freezes there. I tried resetting the bios but it doesn't help. Does anyone know what I should do? Thanks in advance. http://www.flickr.com/photos/51417520@N03/4729633936/ —Preceding unsigned comment added by 72.108.156.138 (talk) 04:51, 24 June 2010 (UTC)[reply]

I built a computer fairly recently that would freeze right at that spot for several minutes before proceeding. Thanks for the bad memories! After trying various things the problem went away when I replaced the DVD drive with another one. I think I changed the SATA cable, as well. Try disconnecting each drive, in turn, and rebooting, and seeing whether that fixes the issue? Comet Tuttle (talk) 05:38, 24 June 2010 (UTC)[reply]
(edit conflict) Your system is likely set up to attempt booting from the hard drive, which has the corrupted Linux installation, before trying to boot from CD. To boot from CD in this case, read your motherboard's manual. It should mention the key you need to press to access the BIOS boot menu (usually F8, F11, or F12; you could try those). That should give you several options; one of those will allow you to boot from CD. Alternatively, you can change the boot device sequence within the CMOS setup utility, placing the CD-ROM drive first (so that the system attempts to boot from CD regardless of whether or not an operating system is installed on the hard drive). PleaseStand (talk) 05:50, 24 June 2010 (UTC)[reply]
Oh, yeah. You should certainly use PleaseStand's suggestion before trying mine. Comet Tuttle (talk) 16:30, 24 June 2010 (UTC)[reply]

Thanks for your help. I tried changing the boot sequence to make the cd-rom as the first priority but it doesn't help. The "verifying dmi pool data..." message comes on too early, before it even boots. I've also tried changing the hard drive, but no matter what I do, it still freezes there. I think the problem might be stored in a different location than the hard drive, since I replaced with a different one and it still doesn't work. —Preceding unsigned comment added by 97.28.3.33 (talk) 07:20, 25 June 2010 (UTC)[reply]

Ugh. One clarification: That whole screen, including the "Verifying DMI Pool Data" message, will always come on before it boots, no matter what drive boot order you are using; it's a BIOS message. It's supposed to only take a second, is the problem. So are you saying you are no longer able to get the computer to boot at all from any device? (Have you tried actually letting it attempt to boot for like 10 minutes?) Could you do another check through the BIOS options to make sure there's nothing that looks like a non-starter? If I were completely unable to boot in this situation after trying several boot devices and changing around the SATA cables to eliminate them as a cause, my next step would probably actually be to try removing all the RAM except for one stick at a time to try to eliminate that as a cause. Comet Tuttle (talk) 17:51, 25 June 2010 (UTC)[reply]
Is your DVD drive connected with a SATA rather than an IDE connection? If so, try putting the DVD drive's SATA port in IDE mode rather than AHCI mode. You might also want to try resetting the BIOS settings to their defaults. PleaseStand (talk) 22:27, 25 June 2010 (UTC)[reply]

I have tried all the solutions except for changing the cables. I will try that next if I still can't get it to work. But this morning I found out if you wait overnight, it solves the Verifying DMI Pool Data. However, it still gets stuck at the next line. http://www.flickr.com/photos/51417520@N03/4737006968/

What happens is it says "Bios autorecovering....." and the dots keep adding until they get to the end of the screen, at which point the computer will restart by itself and begin the cycle again.75.185.120.28 (talk) 01:30, 27 June 2010 (UTC)[reply]

Unfortunately, that sounds like a hardware issue. Motherboards, CPUs, and other components that are bad out-of-the-box are not unheard of. PleaseStand (talk) 21:04, 27 June 2010 (UTC)[reply]
At this point I would try swapping the SATA cables (still trying to boot with only 1 thing connected) and at that point in your situation I'd give up and return the motherboard for a replacement. Comet Tuttle (talk) 01:57, 28 June 2010 (UTC)[reply]

I sent a support ticket to Gigabyte explaining my problem, and they said the motherboard is defective. I'm going to get a replacement tomorrow. Thanks guys.75.185.120.28 (talk) 21:30, 28 June 2010 (UTC)[reply]

Looking for a separate XML Schema file

edit

hello,

I am looking for a separate XML schema of the XML dump "pages-articles.xml" file. It is mentioned in your Website that the XML schemas are defined at the top of the dump file. Since this file is too huge to open in an editor, I'd rather use its schema. How can I have the schema as a separate file?

Thank you.

D.B —Preceding unsigned comment added by 159.84.128.133 (talk) 08:15, 24 June 2010 (UTC)[reply]

Perflib_Perfdata_650.dat

edit

I have this file in my C:windows/temp folder. It will not delete, but is re-made every time I start the computer. I have spent some time on the internet trying to find out what it is. How do I find out what program is using it please? I'm using XP and Firefox. Thanks 92.24.186.235 (talk) 08:56, 24 June 2010 (UTC)[reply]

That's a temporary file used by Performance Monitor. You can find it under Control Panel -> Administrative Tools -> Performance Monitor. However, depending on your version of Windows, by default it will be monitoring a few basic things (memory, processor, network and disk usage on Windows 7 it seems) so having these temporary files isn't unusual. Unless it's using an unusually high amount of disk space, I wouldn't worry about it, it's part of normal Windows operations. ZX81 talk 12:05, 24 June 2010 (UTC)[reply]

turning text into html in JSON

edit

So I finally got my basic javascript Twitter page working. BUT. Twitter sends the "source" element, like what client they use, as a link. But when I just use document.createTextNode on it, it doesn't render it as a link. It just displays the anchor tag text, without rendering it. (for example)

<a href="http://www.tweetdeck.com" rel="nofollow">TweetDeck</a>

Instead of becoming a link, the above just displays as text. The whole thing, <a href> </a> and all. What can I do about this? Thanks, [flaminglawyer] 10:08, 24 June 2010 (UTC)[reply]

I'm not sure what you mean when you say that Twitter sends you data 'as a link', but document.createTextNode will output pure text, so any HTML tags will be lost. Tell us more about what Twitter sends, and I'm sure someone can tell you how to include it into your page. As a quick fix, have you tried to set the sent data directly as an elements innerHTML? Zigorney (talk) 18:48, 25 June 2010 (UTC)[reply]
That was stupid of me. Yes, innerHTML worked. Thanks. [flaminglawyer] 17:58, 26 June 2010 (UTC)[reply]
  Resolved

With the DOS / Command prompt function "cd", is there a way to make it got to the current directory instead of having to re write the script every time I move it to a different folder? 82.43.90.93 (talk) 14:49, 24 June 2010 (UTC)[reply]

This link lists a way to do what it is that I think you want. (I haven't tried it myself. I assume what you want is: your batch file has already done some "cd" statements to change the current directory to be something else, and you then want to "cd" back to the original current directory.) Comet Tuttle (talk) 16:29, 24 June 2010 (UTC)[reply]
Unless I've misunderstood, you can simply use pushd . (the period means set whatever the currently directory is) and popd to jump back to that directory as long as it's in the same session.ZX81 talk 16:43, 24 June 2010 (UTC)[reply]
Hey, that worked for me perfectly. (Though I'm not the original poster.) Comet Tuttle (talk) 18:22, 24 June 2010 (UTC)[reply]
Ah, I thought the OP was asking how to get a batch file to CD to the directory that it's located in. That location is stored in the environment variable %0. But that includes the name of the script, so if this is what the OP wants, s/he should use: cd /d %~dp0 which will CD to the drive and path that the script is stored in. [1] Indeterminate (talk) 00:03, 25 June 2010 (UTC)[reply]

Information about microsoft

edit

can u please provide me with the details about the different components such as finance, production function, marketing, administration, inputs & outputs, research and development and human resource management of microsoft company? —Preceding unsigned comment added by Ashuhaso (talkcontribs) 14:56, 24 June 2010 (UTC)[reply]

You can start with our article Microsoft. Then, here is a copy of Microsoft's most recent Form 10-K annual financial report, which is hard to parse if you haven't read any 10-K's before, but it has a lot of information. Comet Tuttle (talk) 16:26, 24 June 2010 (UTC)[reply]

Captcha is bad

edit

I once saw a website which asked users to prove they're not bots by simply correctly identifying a picture. This looked easier than captcha to me; depending on the words used in captcha, I can end up having to re-prove myself two or three times, but these pictures were easy, I got them first time. Why isn't this used more often? Vimescarrot (talk) 21:20, 24 June 2010 (UTC)[reply]

I imagine it's because a captcha can be generated from any arbitrary word, algorithmically; whereas your describe-a-picture idea requires that when setting up the system, a human must look through many, many pictures (hundreds? thousands?), identify them all, and not make any mistakes. Another robot-identifier that is, I believe, popular in phpBB message boards is to say "What is 2+6?" and have the user enter the result into a text field. This of course is really easy to make a robot solve, if the robot's author has taken the time to make the bot look for a math problem next to a text field. Comet Tuttle (talk) 21:24, 24 June 2010 (UTC)[reply]
Because Mr Spammer quickly collects a massive library of pictures and their valid identifications, obviating the whole thing. To avoid this, captchas (whether read-the-text or not) really need to be mechanically generated. I have seen proposals for ones that show a couple of simple phong-shaded people with a variety of ordinary objects, and you have to answer questions like "what is the object in the man's right hand and the object on the table", and say "umbrella, banana". They vary the lighting and the orientation of everything, and apparently solving this in general is one of the unsolved problems of computer image analysis. -- Finlay McWalterTalk 21:26, 24 June 2010 (UTC)[reply]
All this makes me feel bad because I feel that we're now in a situation where two good-will technologies are competing against each other. We're trying to both enhance automatic visual recognition and hinder it at the same time. Of course these are being done for different purposes, but there's nothing inherently stopping a good-will technology from being used to an evil end. The technology itself won't know a difference. I've even seen Wikipedia being used as a means to help spam evade automatic spam detection. JIP | Talk 21:36, 24 June 2010 (UTC)[reply]
You have a point, but think of it this way; when the image processing improves thanks to nefarious demand, it can then be used for good and a new protection test can be rolled out. It is certainly a cat and mouse game but it's only zero sum if you are in the spam/spam protection business. For the rest of us, we get some nice technology out of the whole deal. Also, see recaptcha for a way to do good while preventing evil. --144.191.148.3 (talk) 17:44, 25 June 2010 (UTC)[reply]
Picture ones can be difficult too. You see a picture of an ear of corn is that "Corn" "Maize" "Grain" "Vegetable" "Plant"? (And too bad if you think corn is a fruit!) APL (talk) 15:01, 25 June 2010 (UTC)[reply]
This is very much tangential, but all of this reminds me of the Army Alpha and Beta tests from WWI... --Mr.98 (talk) 13:57, 26 June 2010 (UTC)[reply]

Meaning of Ubuntu "version?" number?

edit

I have two machines running Lucid Lynx, here are the $ uname -a results:

Linux machine1 2.6.32-22-generic #33-Ubuntu SMP Wed Apr 28 13:27:30 UTC 2010 i686 GNU/Linux

and

Linux machine2 2.6.32-22-generic-pae #36-Ubuntu SMP Thu Jun 3 23:14:23 UTC 2010 i686 GNU/Linux

They are running the same kernel, but the other parts are different. (btw, machine1 was an upgrade, machine2 is a clean install of LUcid) What do the #33 and #36 mean? and can you point to a relevant webpage? Thanks! --Rajah (talk) 23:55, 24 June 2010 (UTC)[reply]

PAE refers to Physical Address Extension. Does machine2 have more RAM than machine1? That might explain the difference. --antilivedT | C | G 05:17, 25 June 2010 (UTC)[reply]
Yes, I know about PAE. Please disregard that for the purposes of this question. Both machines are now up to #36, for example machine1 is now:
Linux machine1 2.6.32-22-generic #36-Ubuntu SMP Thu Jun 3 22:02:19 UTC 2010 i686 GNU/Linux

Ok, doing:

$ grep install /var/log/dpkg.log|grep 36

Turns up:

2010-06-04 19:34:51 status installed linux-image-2.6.32-22-generic 2.6.32-22.36
2010-06-04 19:34:51 status installed linux-headers-2.6.32-22 2.6.32-22.36
2010-06-04 19:34:53 status installed linux-headers-2.6.32-22-generic 2.6.32-22.36

So now, as stated in Software_versioning#Odd-numbered_versions_for_development_releases, they don't just use the fourth number, they also have a fifth number. These pages: https://launchpad.net/ubuntu/+source/linux/2.6.32-22.33 https://launchpad.net/ubuntu/+source/linux/2.6.32-22.36 , etc. seem to indicate it's just a way to version extremely minor changes to the source. --Rajah (talk) 06:54, 25 June 2010 (UTC)[reply]

The part of the version after "#" is taken from the .version file in the source tree where the kernel was compiled. The number in that file is incremented every time a kernel build starts. So it's basically a count of how many times someone ran "make" after untarring the source. This serial number doesn't tell you anything about what's in the kernel. All it does is guarantee that if you recompile, there will be something in the new kernel's version string to distinguish it from the previous one. 98.226.122.10 (talk) 10:17, 25 June 2010 (UTC)[reply]
Thanks! --Rajah (talk) 05:57, 26 June 2010 (UTC)[reply]