Wikipedia:Reference desk/Computing

Welcome to the computing section
of the Wikipedia reference desk.

skip to bottom

Select a section:

• Computing
• Entertainment
• Humanities
• Language
• Mathematics
• Science
• Miscellaneous
• Archives

Want a faster answer?

Main page: Help searching Wikipedia

   

How can I get my question answered?

• Select the section of the desk that best fits the general topic of your question (see the navigation column to the right).
• Post your question to only one section, providing a short header that gives the topic of your question.
• Type '~~~~' (that is, four tilde characters) at the end – this signs and dates your contribution so we know who wrote what and when.
• Don't post personal contact information – it will be removed. Any answers will be provided here.
• Please be as specific as possible, and include all relevant context – the usefulness of answers may depend on the context.
• Note:
  • We don't answer (and may remove) questions that require medical diagnosis or legal advice.
  • We don't answer requests for opinions, predictions or debate.
  • We don't do your homework for you, though we'll help you past the stuck point.
  • We don't conduct original research or provide a free source of ideas, but we'll help you find information you need.

Ready? Ask a new question!

How do I answer a question?

Main page: Wikipedia:Reference desk/Guidelines

• The best answers address the question directly, and back up facts with wikilinks and links to sources. Do not edit others' comments and do not give any medical or legal advice.

See also:

• Teahouse
• Help desk
• Village pump
• Help manual

November 16

windwl32.rom

Hello Everyone,

   Upon booting up (or restarting), my computer now displays the message "There was a problem starting windwl32.rom The specified module could not be found." This missing module does not seem to cause any problems thereafter. What is this module, does its absence adversely affect the operating-system, and what is the appropriate remedy?

   Thank you - any insight is appreciated.

That file is not a necessary part of the standard Windows system. A web search indicates that the file "windwl32.rom" is associated with a malware infection. You may want to completely reinstall your operating system. At the very least, consider running some malware-removal software. Nimur (talk) 01:06, 16 November 2010 (UTC)

I'd stab that it's Vundo. A strong rinse in antivirals could fix it. ResMar 03:56, 16 November 2010 (UTC)

I have performed a full scan (which took three hours) of my computer with McAfee Security Center, and it reported "No Issues Detected: During your scan, McAfee did not detect viruses, spyware, or other threats. Remember, you can run a full scan any time to thoroughly check your PC for threats." Nonetheless, the message "There was a problem starting windwl32.rom The specified module could not be found." is still displayed upon booting up. How do I get rid of this malware without wiping my C: partition and reinstalling my operating-system; this is not possible because I do not have the reinstallation disc(s). Rocketshiporion♫ 11:52, 16 November 2010 (UTC)

In the past, I've had issues where the virus is removed, but the registry entries that started it at boot-time are still there. Virus detectors tend not to pick that up, presumably because they look for the virus itself, not its lingering side-effects.

That sounds like what you're describing. If the virus itself was still there, you wouldn't get the 'module not found' error, it would find the module, start it, and do nefarious things to your computer.

If you're comfortable poking around in your computer, you might use the registry editor tp search for all references to "windwl32.rom" and get rid of them. You do that at your own risk, though. You can seriously break your computer with the registry editor. APL (talk) 15:26, 16 November 2010 (UTC)

Perhaps running MalwareBytes Anti Malware will help? Maybe changing the executable name may work (if the malware is still present, that is) General Rommel (talk) 21:24, 16 November 2010 (UTC)

   I ran Malwarebytes Anti-Malware, and then used regedit.exe to search for remove one remaining reference to windwl32.rom (which I discovered under HKEY_LOCAL_MACHINE\SECURITY). My computer no longer displays the "There was a problem starting windwl32.rom The specified module could not be found." message upon boot-up. So, thank you to everyone, and in particular to Nimur (for informing me that windwl32.rom is not part of a standard Windows system), APL (for suggesting the use of the registry-editor) and General Rommel (for pointing me to Malwarebytes' Anti-Malware). Rocketshiporion♫ 22:23, 17 November 2010 (UTC)

PPP connection with Ubuntu

I am trying to connect to the Internet through a PPP connection.

I thought I had everything right and gnome-ppp appears to establish a connection, but no browser seems to recognize the connection. I have tried changing the proxy preferences of the browsers but no luck so far.

I am using a cell-phone as USB modem and I have connected with this same cell-phone before (but under Windows).

My firewall recognizes the PPP connection. (The problem is not the firewall, even if I turn it off, the problem is not solved).

Thanks in advance for any help.

This is what I get from gnome-ppp's log:

Extended content

WVCONF: /root/.wvdial.conf GNOME PPP: Connecting... GNOME PPP: STDERR: --> Ignoring malformed input line: ";Do NOT edit this file by hand!" GNOME PPP: STDERR: --> WvDial: Internet dialer version 1.60 GNOME PPP: STDERR: --> Cannot get information for serial port. GNOME PPP: STDERR: --> Initializing modem. GNOME PPP: STDERR: --> Sending: ATZ GNOME PPP: STDERR: ATZ GNOME PPP: STDERR: OK GNOME PPP: STDERR: --> Sending: ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 GNOME PPP: STDERR: ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 GNOME PPP: STDERR: OK GNOME PPP: STDERR: --> Modem initialized. GNOME PPP: STDERR: --> Sending: ATM1L3DP*99# GNOME PPP: STDERR: --> Waiting for carrier. GNOME PPP: STDERR: ATM1L3DP*99# GNOME PPP: STDERR: CONNECT GNOME PPP: STDERR: --> Carrier detected. Starting PPP immediately. GNOME PPP: STDERR: --> Starting pppd at Tue Nov 16 00:04:30 2010 GNOME PPP: STDERR: --> Pid of pppd: 6981 GNOME PPP: STDERR: --> Using interface ppp0 GNOME PPP: STDERR: --> Authentication (CHAP) started GNOME PPP: STDERR: --> Authentication (CHAP) successful GNOME PPP: STDERR: --> Terminate Request (Message: "No network protocols running" ) GNOME PPP: STDERR: --> Terminate Request (Message: "No network protocols running" ) GNOME PPP: STDERR: --> Disconnecting at Tue Nov 16 00:05:08 2010 GNOME PPP: STDERR: --> The PPP daemon has died: A modem hung up the phone (exit code = 16) GNOME PPP: STDERR: --> man pppd explains pppd error codes in more detail. GNOME PPP: STDERR: --> Try again and look into /var/log/messages and the wvdial and pppd man pages for more information.

Mr.K. (talk) 12:29, 16 November 2010 (UTC)

Do look at /var/log/messages like it said. There's probably more information there. It seems to be failing in the IPCP phase (which is where the 2 endpoints negotiate IP addresses) but from the incomplete log above we can't see why. Also you could try adding "debug" to /etc/ppp/options and if you're going to follow up, also mention the name of your service provider. 67.162.90.113 (talk) 08:25, 17 November 2010 (UTC)

OK. Thanks for the help so far. At the bottom is the information from /var/log/messages. You seem to be right when you say that the problem is the IPCP. Actually I do not get any IP number, even if gnome-ppp sets the status as connected and report that some data (a couple of KB) is being transmitted.

My service provider is the Spanish branch of Orange.

Extended content

Nov 17 12:41:01 mrk-laptop pppd[6413]: pppd 2.4.4 started by root, uid 0 Nov 17 12:41:02 mrk-laptop pppd[6413]: Using interface ppp0 Nov 17 12:41:02 mrk-laptop pppd[6413]: Connect: ppp0 <--> /dev/ttyUSB0 Nov 17 12:41:02 mrk-laptop pppd[6413]: CHAP authentication succeeded Nov 17 12:41:02 mrk-laptop pppd[6413]: CHAP authentication succeeded Nov 17 12:41:02 mrk-laptop kernel: [ 1201.019309] PPP BSD Compression module registered Nov 17 12:41:02 mrk-laptop kernel: [ 1201.177566] PPP Deflate Compression module registered Nov 17 12:41:34 mrk-laptop pppd[6413]: IPCP: timeout sending Config-Requests Nov 17 12:41:40 mrk-laptop pppd[6413]: Connection terminated. Nov 17 12:41:40 mrk-laptop pppd[6413]: Modem hangup Nov 17 12:41:40 mrk-laptop pppd[6413]: Exit.

for enquiry

dear madam/sir:

my question is if i want to get any information from wikipidia , what procedure will follow to the required information,

eg: how can i get the list of I.T parks in U.A.E from wikipidia —Preceding unsigned comment added by 122.166.12.77 (talk) 16:20, 16 November 2010 (UTC)

Hi. We have a list of technology centers worldwide; but the only one listed in UAE is Dubai Internet City. That article mentions a few others, including Ras Al Khaimah IT Park. We may have information on other technology centers in UAE, but it doesn't look like they're organized in a single list yet. Nimur (talk) 17:08, 16 November 2010 (UTC)

If your question is about how to locate information on Wikipedia; you can either type keywords into the Search-Box (which is located on the top-right corner of every page in Wikipedia) or use Wikipedia's Search page. Rocketshiporion♫ 22:29, 17 November 2010 (UTC)

Creating a wireless network using an aircard

I have an aircard from Verizon Wireless. I also have a wireless printer and a Wii. I was under the impression that I could get a router to plug my aircard in to, to create a wireless network that I can connect my computer, printer and Wii to. The two options I have come up with are the Verizon MiFi and a Cradlepoint. I have read that the MiFi is not worth the charges when going over your minutes, which apparently happens very frequently. But I do not know anything about the Cradlepoint, which has been a very popular suggestion. Can anyone help me find out about the Cradlepoint, which one I should get, or something else that can give me a wirless network without breaking my wallet? Thanks in advance!Amstokes1 (talk) 18:31, 16 November 2010 (UTC)

Your friend asked the question below, and I answered it there - but the article on celluar routers might help. A MiFi will do what your aircard + cellular router could do, with less components. I'm not an expert on Verizon's data plans, but if you have unlimited data it should not be an issue with either device. If you are out of contract on your aircard, you may be able to get a MiFi for cheap or free with a contract renewal. Do keep in mind that DSL or cable modem Internet access will likely be cheaper in the long run, and provide you with faster speeds and greater reliability. It's not mobile like the aircard or MiFi, though. coreycubed / talk 19:14, 17 November 2010 (UTC)

making "sed" match only once

Resolved

I'm trying to make a single change in a text stream using sed; that is, I want one change for the whole stream, not one change per line. So I've been running:

printf "hello jello\nsmello\n" | sed "s/ll/y/1"

and getting

heyo jello
smeyo

whereas I want

heyo jello
smello

Is there a way to make sed make a substitution exactly once, or alternatively to make it stop treating each line as an independent thing? 87.112.174.140 (talk) 18:34, 16 November 2010 (UTC)

Add "0,/pattern" to the beginning of the sed command, as in: printf "hello jello\nsmello\n" | sed "0,/ll/s/ll/y/1" -- kainaw™ 19:31, 16 November 2010 (UTC)

That's just the ticket. Thanks for your help. 87.112.174.140 (talk) 19:46, 16 November 2010 (UTC)

When did "I think I have I a virus/malware/spyware etc" = OMFG REINSTALL YOUR ENTIRE OPERATING SYSTEM FROM SCRATCH BEFORE YOUR PC EXPLODES!

I've noticed at least a couple of answers recently where the poster has reported a possible infection with some kind of rogue program and the first or an early answer is to advise them to reinstall their operating system. Unless you are pretty tech savvy, this is NOT a trivial procedure - it's time consuming, not everyone has the original installation discs and they then have to download every security fix and patch and probably reinstall all their software. Surely this should only be considered as a LAST resort when all other attempts to remove the infection have failed? Most normal computer users will never install or reinstall an operating system EVER - they are liable to simply replace the computer with a new one with the same or a later version of Windows. Why would anyone advise this as an initial response to the problem? Exxolon (talk) 22:40, 16 November 2010 (UTC)

Some of the viruses nowadays are pretty damn tenacious. Getting rid of them can almost literally be as much trouble as reinstalling. HalfShadow 22:42, 16 November 2010 (UTC)

I think its because of one or two fanatics here. I disagreed with that approach. I'm tempted to delete the viruses article - the one with a nuclear plume - and start again, but I bet it would get reverted by its 'owner'. 92.15.16.149 (talk) 23:28, 16 November 2010 (UTC)

Because it's a cure all way (99.999% of the time). Also it gives them a chance to make sure the computer won't die another time. And also makes sure that the virus is fully cured, some programs may not fully remove the virus and traces may remain. General Rommel (talk) 23:51, 16 November 2010 (UTC)

Its a cure, like amputating your head wopuld be if you have a headache. 92.28.250.11 (talk) 11:04, 17 November 2010 (UTC)

In What Colour are your Bits, (which is actually an essay only tangentially related to malware), there is a pretty profound discussion of why we trust a "cleanly installed" operating system, but we do not trust a compromised operating system. "The trouble is that because any of our files might be infected including the tools we would use to test for infection, we can't reliably compute the "is infected" function,..." This problem is about establishing a base-level of certainty about what is running. When a system is compromised, we don't know how badly it is compromised. Basic features - like "copy file" and "delete file" may be replaced with illicit versions by the mal-ware. By definition, using any tool to clean a malware infection is trusting that the malware has not taken certain defensive actions. To be on the safe side, we presume that any compromised system is completely compromised, and can not be trusted to assist in "cleaning" itself. I don't think this is heavy-handed - I think it is a correct assessment of the way malware works in 2010. Can you, for instance, guarantee that your shell32.dll is the proper one? Suppose you even want to use a technical mechanism to verify it: let's consider the MD5 hash of certain critical system files. Do you know their MD5 hash off the top of your head? Can you trust that any program(s) you use to compute or verify any MD5 hashes will operate properly when you know your system has already been compromised? You can't - so if you feel comfortable "trusting" the compromised system, you're taking a huge risk. Nimur (talk) 06:47, 17 November 2010 (UTC)

As the OP points out, what do the fanatics suggest that a user does who has not got any OS disks - probably most people? 92.28.250.11 (talk) 10:59, 17 November 2010 (UTC)

If you HONESTLY want this from the opposite point of view... Whenever I visit anyone (friends, family, enemies...), or even whenever I get a phone call, it is the same scenario. "Hi, how you doing..." That is just a formality. It doesn't take long before the conversation turns to, "Hey, Mr. PhD of Computer Science, make my computer stop doing (fill in the blank)." Every single time, this error just appeared. The user did nothing to the computer at all. Did he install anything? No. Was he running any programs? No. Was he even anywhere near the computer within the last few days? No. Has he ever turned on the computer since he took it out of the box? No. The user will not admit to ever touching the computer, but wants me to fix some random error. I know that he installed something stupid, but he won't tell me what it was. So, I don't want to waste my time screwing around with his computer. I tell him it is a Windows thing. Fdisk, format, reinstall. What? No install disks? That's a Windows thing. Go get some. Then, Fdisk, format, reinstall. What? You want to save your files? Losing files is a Windows thing. Fdisk, format, reinstall, and forgot about all those files you lost. Of course, it isn't a "Windows thing." It is a "user thing." But, after 20+ years of "fix my computer" requests, I simply don't care to poke around through Windows to figure out what the user did so I can try to figure out how to fix it. So, I do not answer the "how do I make Windows stop doing (fill in the blank) questions here." But, I do perfectly understand those who answer those questions with "Fdisk, format, reinstall." -- kainaw™ 13:29, 17 November 2010 (UTC)

Under most circumstances, I like to recommend the standard virus scan + any virus specific removal tools + one other malware scanner (ideally run off a CD/flash drive). However in some cases, it really is easier to simply back up what's important and start from scratch (as the last virus question seemed to indicate). If scans and removal tools have already been run and there are still signs of the virus or if it left junk around, the process of cleaning up after it can often be more time consuming and painful than simply reinstalling the OS and all your programs. It should never be the first thing to try, but once there are signs of the current operating system being too far gone to bother cleaning it, then it's a valid suggestion. 206.131.39.6 (talk) 15:42, 17 November 2010 (UTC)

I think the problem is if you have malware in the first place and then have to ask here for help, there's a fair chance properly removing the malware is beyond your level of expertise. Reinstalling isn't trivial although many OEM computers come with some sort of partition or media which may make it easier. Reinstalling software is a pain, but if your the person who's got malware because you try everything and sundry it may be best to start fresh and if you're not there's a far chance software consists of browser and associated plugins, perhaps an email client, perhaps some Office software and perhaps a media player. As others have said, even if you do have a high level of expertise guaranteeing theres no trace arguably is very difficult although depending on the malware you may be confident enough for your purposes. Nil Einne (talk) 16:05, 17 November 2010 (UTC)

As for what do the fanatics suggest that a user does who has not got any OS disks — I disagree with the premise, and also with the false "OMFG" premise of the original poster's section header; I am not a "fanatic" about this, but personally I prefer to be sure that the malware is gone, and the only sure way, as pointed out above, is to reinstall from scratch, because it is impossible for a user to know what exactly has been compromised, despite the best intentions of anti-malware software writers. As for the answer to the question, it seems obvious that if a user doesn't have the OS disks or the equivalent partition (as all HP computers have these days, for example), then the user has no choice but to gamble, and just use anti-malware software, and hope for the best. Comet Tuttle (talk) 18:12, 17 November 2010 (UTC)

The Wikipedia article about viruses ought to suggest doing scans etc first, not just Armagedon. 92.15.28.182 (talk) 18:15, 17 November 2010 (UTC)

Wikipedia is not a manual for removing viruses. So, it shouldn't suggest any method for removing them. The article on viruses should explain what they are and nothing more. -- kainaw™ 18:30, 17 November 2010 (UTC)

92 is referring to the RD "Viruses" FAQ at Wikipedia:Reference desk/Computing/Viruses, which is as unofficial as anything else on Wikipedia, and which is intended to help us answer the very frequent question encountered here, "What do I do to get rid of my malware", without us answerers forgetting things, and without having to write an entire page each time someone asks this question. To 92: I agree; please help us by improving the FAQ. But you haven't rebutted any of the arguments in this thread that detail how scans may not work, depending on how your system has been compromised. Comet Tuttle (talk) 19:41, 17 November 2010 (UTC)

See the third comment down. 92.24.187.23 (talk) 13:30, 18 November 2010 (UTC)

Why is my system so slow with 2D DirectX graphics?

I recently updated/rebuilt my computer, here are the basic system specs of the new system:

• Asus M4A87TD board (upgraded from a crappy Elitegroup MCP78M)
• AMD Phenom II X2 545 (upgraded from AMD Athlon II X2 240)
• 4GB RAM, DDR3 @1333 (upgraded from 2GB DDR2, 256MB of which was used by the onboard graphics card)
• Geforce 250 GTS, using Nvidia's 260.99 driver (upgraded from the crappy onboard Geforce 8100)
• WinXP SP3 (fresh install, no change otherwise)

I know it's still not exactly a high-powered gaming rig, but it's enough for my needs. As expected, I experienced huge performance increases in 3D games (tested with Portal, GTA4 and Mafia 2, all of which run much smoother now than with the old system). What I don't get is that 2D DirectX graphics are markedly slower than on the old system; it's especially noticeable with my screensaver (I use JWZ's Matrix screensaver, a simple 2D display scrolling down letters in various shades of green) and with RPGmaker XP games - I'm toying around with making games with RPGmaker, the engine is pretty crappy and starts lagging heavily as soon as there is a bit going on on the map so a good part of making games in RPGmaker consists in coming up with clever anti-lag measures anyway, but the lag issues have become really unbearable on the new system. I've even noticed some lag when scrolling large maps in the editor. So...does anybody have an idea what's going on here? I'm aware 2D graphics do not benefit from the Geforce's 3D acceleration, but with twice the memory and a slightly faster processor I would have expected 2D performance to increase at least slightly, or at the very least stay the same. -- Ferkelparade π 23:13, 16 November 2010 (UTC)

November 17

Autorun problem in Windows 7

The following is saved under autorun.inf in my flash drive's root directory. It successfully changes the flash drive's name, but the READ ME option doesn't show up on the autorun menu when I insert the flash drive. What's the problem?

[autorun]

Open=READ ME.bat

Action=READ ME

Label=[my name]

--75.33.217.61 (talk) 02:38, 17 November 2010 (UTC)

I'm not entirely sure that this will fix the problem, but you could try enclosing the values in quotation marks, like this

[autorun]
Open="READ ME.bat"
Action="READ ME"
Label=[my name]

Rocketshiporion♫ 02:45, 21 November 2010 (UTC)

I think the problem is that Microsoft disabled autorun for USB drives and other re-writeable media as it was often used to spread viruses. See this: http://www.microsoft.com/whdc/connect/usb/usbfaq.mspx

Wireless network for a Verizon phone

A friend of mine asked me:

I have a Verizon Wireless Aircard, I have a wireless printer but apparently it does not connect to aircards... I have read about something called a Cradlepoint that you plug the aircard into and it creates a wireless network that I could connect my computer, printer and Wii to... Know anything about it.

What we mainly are interested in is the best way to create a wireless network by buying hardware and not paying a monthly fee. What's the best way? Thanks --Wonderley (talk) 02:52, 17 November 2010 (UTC)

Hmm, bizarre. I see my friend posted something here. I did a search on several keywords before adding this post with no hits. Feel free to answer either of us. —Preceding unsigned comment added by Wonderley (talk • contribs) 03:03, 17 November 2010 (UTC)

You've got two different kinds of wireless going on here. The Verizon Wireless aircard is providing access to their EV-DO mobile broadband over cellular networks. The rest of the wireless technology you described is for local networking - a wireless LAN, or "Wi-Fi". This will network your devices, but you will still need a source of Internet access provided by either a regular DSL or cable modem, or via cellular networks like the Aircard you already have. If you want to continue using the Aircard to provide Internet access to all of your devices, then you need a cellular router. A modular cellular router will take the Aircard and create a new, local wireless network that your other devices (such as your wireless printer and your Wii) will be able to connect to.

Personally, I'd recommend getting a dedicated connection via DSL or cable - speed and reliability will be greatly improved. They're not dependent on cellular coverage. coreycubed / talk 19:06, 17 November 2010 (UTC)

How do I restore all extensions and tabs on my Google Chrome?

After restarting from my computer freezing up, all my Chrome extensions and tabs were gone. The bookmarks were there, but the History was erased as well, so I can't reload the 38 tabs manually either.

My SessionBuddy has a saved session of all my tabs, but first I need to figure out how to put SessionBuddy back on there. (I downloaded it, so the folder is someplace.) Then I'd also appreciate putting back all the other extensions I've had. Please help me get them all back. Thanks. --Let Us Update Wikipedia: Dusty Articles 08:42, 14 November 2010 (UTC)

What disk formats can PS3 mount through USB port?

I know FAT32 works with everything. But FAT32's 2GB file size restricts the PS3 as a non-networked media playback device. Are there any other disk formats that the PS3 will recognize through the USB port? --70.167.58.6 (talk) 13:33, 17 November 2010 (UTC)

From the PS3 manual: "Note however that the disk must be formatted in the FAT32 file system to be recognized by the PS3™ system." -- kainaw™ 14:18, 17 November 2010 (UTC)

FAT32 doesn't have a 2GiB file size limit. As our article mentions, the limit is 4GiB-1B Nil Einne (talk) 15:56, 17 November 2010 (UTC)

How old is that manual? Does it cover the latest 3.50 update? --70.167.58.6 (talk) 16:42, 17 November 2010 (UTC)

Yes. For a question about using PS3, I would go directly to the manual. -- kainaw™ 17:15, 17 November 2010 (UTC)

Clock in RISC OS 3.7

I found this screenshot of the clock in RISC OS 3.7. What happens if you scroll down? --Andreas Rejbrand (talk) 15:06, 17 November 2010 (UTC)

Below the clock, it also shows the time as numbers, ie: 11:25:32. That is being cut off by the window, which is why there is a scrollbar. -- kainaw™ 19:41, 17 November 2010 (UTC)

Resolved

High capacity text editor?

I routinely need to hand edit some large (500MB) Maya ASCII files. Are there any text editors that are still nimble with such a heavy load? Ideally, I'd like a Mac OS X editor, but I can run Windows 7 in boot camp. --70.167.58.6 (talk) 16:40, 17 November 2010 (UTC)

Here and here are pretty good discussion of this for primarily Windows programs. Basically the consensus seems to be that the "best" way for editing such files is to use something like vim (gVim) or emacs, neither of which are very user friendly (which is code for me saying "they require you to do a lot of work to learn how to use them, like many Unix-y programs"), but can handle arbitrarily large files without a real performance hit. Standard "notepad replacement" programs like Notepad++ and TextWrangler or BBEdit either have real difficulties with files that size or simply cannot handle them because of RAM buffer limitations. VEDIT gets high marks from one poster. --Mr.98 (talk) 17:06, 17 November 2010 (UTC)

emacs should do it. Note that if you just need to search and examine a very large file, less is brilliant, and if you need to make some routing change (like search-and-replace some text to another) then sed can allow you to automate a lot of such tasks. -- Finlay McWalter ☻ Talk 17:09, 17 November 2010 (UTC)

I use EditPadPro [1] for a variety of text editing applications and am very happy with the facilities it provides. It's not free, but does some good stuff. I've just created a 1.4 Gig text file using it, and it takes a while to open and save it, but otherwise seems to handle it pretty well (with the odd out-of-memory error while doing all the copy-paste to get a file of that size). --Phil Holmes (talk) 18:14, 17 November 2010 (UTC)

There's a stackoverflow question on this. Topgun, written in MASM should be able to handle large files...I've tested it on 2gb file in Win7. Just be aware that for most editors, topgun included, you'll need have enough free ram to open the file(the file is loaded into ram).Smallman12q (talk) 15:15, 18 November 2010 (UTC)

Sorting by date modified

Working on an iMac G5, OSX, version 10.5.8. I have a ton of Word documents in various files. When I open a file directly in a window, the documents can be listed by the fields: "Name", Date Modified", "Size" and "Kind". However, when I perform a search, the list of Word documents that match the search entry only have the fields: "Name", "Kind and "Last Opened". I want to be able to see and sort by Date Modified here too, which is much more useful to me than is "last opened". Can anyone explain to me, step-by-step, how I would add this function? Thanks.--108.27.105.73 (talk) 18:53, 17 November 2010 (UTC)

Step-by-step:

1. Search for the files (in Spotlight > Show All, or just in the Finder window).

2. While the "Searching 'This Mac'" window is highlighted, go to View > Show View Options (Apple+J).

3. Select "Date Modified".

Should do it? Works on mine, though I'm using 10.6. It seems to make any columns added that way permanent features to Spotlight searches.--Mr.98 (talk) 19:14, 17 November 2010 (UTC)

Thanks much for the explanation! I was able to follow perfectly. However, when I go to show view option, the menu that appears says "There are no view options for the "searching "this Mac"" window." Any ideas?--108.27.105.73 (talk) 19:32, 17 November 2010 (UTC)

A further update. There are multiple macs in the office. I just tried your instructions with a different mac and it worked beautifully. Still stuck on mine though.--108.27.105.73 (talk) 19:37, 17 November 2010 (UTC)

You might look at the version (Apple > About This Mac) on the one it works with. It might be a 10.6 (Snow Leopard) only addition and not present in 10.5 (Leopard). --Mr.98 (talk)

I Am Still SoapFiend

Okay, so, I am still SoapFiend, but some &^%$#@# genius decide to hack into my ^%$#@!@! user account on this website and I can't even read the *&^%$## watchlist like I always do because of the motherz*&^^%$% "session hijacking" bull*&^%. --64.123.104.83 (talk) 20:58, 17 November 2010 (UTC)

All I want to do is to check into my user page and look at my watchlist and see if there are some changes in my articles related to television and television actors and actresses. How do I get my user account back in order for me to do that again? --64.123.104.83 (talk) 20:58, 17 November 2010 (UTC)

You need to be more clear. Are you complaining that you forgot your password, or are you complaining that your account was blocked? Comet Tuttle (talk) 21:42, 17 November 2010 (UTC)

I think he means that he wants to trick an administrator into naively removing SoapFiend's indef-block. APL (talk) 03:52, 18 November 2010 (UTC)

• "How to get your user account back" is pretty well explained here. Arakunem^Talk 15:22, 18 November 2010 (UTC)

You don't need to be unblocked to look at your watchlist AFAIK. You do need to know the password, without a committed identity, unless you are able to reset your password (in which case you don't need to ask anyone) you're SOL AFAIK if you don't. Nil Einne (talk) 21:24, 18 November 2010 (UTC)

See also WP:GOTHACKED and WP:BROTHER.—Emil J. 16:38, 18 November 2010 (UTC)

Coding help

Hi there. The following is a dreadful piece of code, but it works:

void SplitNumber(double dValue, double* dMantissa, double* dExponent)
{
    char szValue[32];
    char* pTemp;

    sprintf(szValue, "%e", dValue);
    pTemp = strchr(szValue, 'e');
    *pTemp = '\0';
    pTemp++;
    *dMantissa = atof(szValue);
    *dExponent = atof(pTemp);
}

Does anyone have any suggestions for a less-dreadful piece of code that still works? 80.254.147.84 (talk) 21:51, 17 November 2010 (UTC)

We need to ask - how portable do you want this code to be? The C language does not specify what format a float or double should take; but if we can assume you're running on a "normal" computer that supports Single precision floating-point format and double precision floating-point format according to IEEE 754, then we can use a bit of bit-masking. Do you need help writing code to apply a bitmask for the exponent and mantissa as illustrated in our articles? Be sure to take note of the special-cases of exponents. Nimur (talk) 22:27, 17 November 2010 (UTC)

Check out frexp(). It's standard. Example usage (in C++, but it's the same)..--Sean 22:34, 17 November 2010 (UTC)

(That's much easier than my system!) But note that frexp() returns results in power-of-two - whereas the original code returned results as power-of-ten - so a bit of intermediate conversion will be required. Nimur (talk) 22:56, 17 November 2010 (UTC)

How about:

void SplitNumber(double Value, double* Mantissa, double* Exponent)
{
    if (Value == 0) {
      *Exponent = 0;
      *Mantissa = 0;
    } else {
      *Exponent = floor(log10(fabs(Value)));
      *Mantissa = Value / exp10(*Exponent);
    }
}

(I have removed the "d" prefixes, which in a modern system accomplish nothing except to make the code less readable.) Looie496 (talk) 00:03, 18 November 2010 (UTC)

Thanks for your help. If Nimur's intermediate conversion will need logs anyway, Looie496's solution looks good. One more question - is it OK just to check if dValue equals zero, or should I check that fabs(dValue) is greater than some critical figure? It won't be much less than about 1E-12 in reality, but it's quite likely to be zero. For portability, it just needs to run on Windows 2K and XP - not Vista or anything newer, or W98 or anything older. 80.254.147.84 (talk) 20:29, 18 November 2010 (UTC)

In this case checking against 0 is all you need. All other (finite) values have representable logarithms. --Tardis (talk) 22:01, 18 November 2010 (UTC)

Note that using Looie's function with an input of 99.99999999999998 will probably yield an exponent of two and a mantissa that's less than 1, while the original function will probably (correctly) yield an exponent of 1. This happens because the result of log10 (which is very slightly less than 2) is rounded to the nearest double (which is 2) before being passed to floor (which returns 2, since its argument is exactly 2). Unless you need speed at the expense of accuracy, I would actually suggest using the original function, except replacing "%e" with "%.18e" or something of that sort so that all input precision is preserved. Solving this problem correctly for all inputs is surprisingly difficult, and using the C library's implementation may be your best bet. (Unfortunately, I think that my suggested solution with "%.18e" will sometimes yield an improperly rounded value for the mantissa, and I see no easy way to fix that. For nonzero input it should never yield a mantissa outside the range [1,10], but it may yield a mantissa exactly equal to 10.) -- BenRG (talk) 01:00, 19 November 2010 (UTC)

Thanks again. Speed isn't important and for what I want six digits of precision will be more than enough, but I can add a precision parameter to make the routine more general. (That makes it

    sprintf(szValue, "%.*e", iPrecision, dValue);

doesn't it?) 80.254.147.84 (talk) 20:17, 19 November 2010 (UTC)

Why lose any precision? Rounding is for output, never for processing. --Tardis (talk) 23:17, 19 November 2010 (UTC)

We can fix that by just testing: if(*Mantissa<1) {*Mantissa*=10; --*Exponent;} else if(*Mantissa>=10) {*Mantissa/=10; ++*Exponent;}. --Tardis (talk) 16:24, 19 November 2010 (UTC)

Downloading Silverlight Video?

What's a good way to download embedded silverlight video (preferably free of crappy adware-laden software packages)

Thanks in advance,

PerfectProposal 23:01, 17 November 2010 (UTC)

You will have to make sure the video is DRM free, otherwise you will not be able to play it. Graeme Bartlett (talk) 05:42, 18 November 2010 (UTC)

November 18

Post compile array size definition in Pascal

I'd like to create an array such as the type:

truthtable : array [1..n,1..x] of boolean;

in pascal after having used readln(n), and x:=n-1 to get the variables.

As far as I can tell the compiler only allows constants in the definition. Excluding creating a block of memory at run time and not using an array is there a work around ? 83.100.237.223 (talk) 01:41, 18 November 2010 (UTC)

You should use a dynamic array:

var
  truthtable: array of boolean;
begin
  SetLength(truthtable, n, x);
  // Now you got truthtable as an array[0..n-1, 0..x-1] of boolean
end;

--Andreas Rejbrand (talk) 08:29, 18 November 2010 (UTC)

Thanks . Actually worked with

var
 truthtable : array of array of boolean;
 ...etc

83.100.237.223 (talk) 18:17, 18 November 2010 (UTC)

Resolved

Programming language

So I'm thinking about learning a more "hardcore" programming language.

I work on a Mac (so no Visual Studio products). I am pretty proficient with PHP, Javascript, VBScript, and Actionscript. I can meddle with VB.NET but I haven't used it as much as I could. I pick up things pretty quickly and have a fairly good programming mind. I know the basics of programming logic and object oriented programming. Of the languages I know, I find PHP the most straightforward and VBScript the least straightforward. What I like about PHP is I find it very easy to translate the logical structure in my head into the code on the page, without having to jump through too many jargony hoops, if that makes sense.

What's a good next step? I'd like whatever language I use to not involve too much meddling with the nuts and bolts of computer memory. I'd like it to be cross-platform enough to develop on my Mac, but to make programs that could theoretically be ported to PCs. I'd like it to be able to create GUI programs that look pretty good. The programs it make ought to be easily distributable as binaries.

I'm not 100% sure what I'd use it for at this point — I make lots of little things, and find that obviously having the capacity to use the language inspires me to have ideas as to what to make with it. My general model will probably be to get a "how to learn language X" book and work through it first, whatever the language.

I'm leaning towards Java. It seems to allow more portability than most others, seems to be relatively flexible and popular, and seems like it would open up opportunities for using it, say, in conjunction mobile apps.

Any other suggestions, thoughts, warnings? I understand this is entire subjective. --Mr.98 (talk) 01:57, 18 November 2010 (UTC)

If you are programming applications for "PC users," I would vote for Java as the best all-around power and tradeoff between portability, functionality, performance, and (this is critical), quality of free tools and documentation. If you decide to program embedded devices at a low level, there's no substitute for C. But nowadays, many "mobile" devices like phones and ipods provide platform-specific programming environments - so you often have to target a specific brand/technology and use that set of tools. Nimur (talk) 02:12, 18 November 2010 (UTC)

At first sight there doesn't seem to be much choice for languages with good GUI support, (no .NET on the mac). As you don't want nuts and bolts C++ sounds like a non starter - the main difference is the need to do memory management yourself - depending on the scope of your programs that might not be a big issue, or you might find you enjoy the task. There are some other options though, notably Ada and Lisp which can be used with the Qt widget toolkit amongst others.

I guess that when you say 'hardcore' you mean faster. Ada is fast, and a nice easy language to read, unfortunately it doesn't have a total feature set on either GTk of Qt widget toolkits according to the wikipedia articles. Lisp is well supported (in terms of GUI stuff etc) and is a totally different kettle of fish, it's just something you'd have to look at yourself to see if you think it'll be useful to you. There are also some other ALGOL descendents that have 'rapid gui development tools' - such as pascal + Lazarus (software) or other gui builders. Pascal is a bit old though, I'm not sure how well it would adapt to a network type setup. None of these would be of much use on a mobile device though as far as I know.

The page Comparison of integrated development environments might be useful - not all the IDE's listed will have the features you want - but it's a good place to start browsing.83.100.237.223 (talk) 03:35, 18 November 2010 (UTC)

I'm not really sure what would constitute a "hardcore language"...especially since you don't want to learn pointers. Java/.Net would be pretty good... REALbasic also has cross-platform support. You might also want to see Comparison of C Sharp and Java.Smallman12q (talk) 15:10, 18 November 2010 (UTC)

I just want to note that I've played with REALbasic before and found it too primitive for my needs, and not cross-platform enough where it counted. It's like VB6 that works on a Mac, basically, which will get you so far but what I'm hoping for is something that lets me not rely on third parties to write plug ins and things like that so heavily. There are things it can't do at all (like adequate PDF support, or even the ability to use standard PDF viewer plugins) that are kind of musts for anything I want to invest a lot of time in. --Mr.98 (talk) 01:52, 19 November 2010 (UTC)

C, Perl. Java™ isn't remotely hardcore, it's what they teach kids at university instead of Pascal, because after a long time people figured out Pascal wasn't hardcore, so they had to switch to something else, and they'll have to again, eventually. ¦ Reisio (talk) 19:11, 18 November 2010 (UTC)

Thanks for the suggestions. I think I'm going to go with Java, or at least give it a gander. I appreciate your taking the time to try and make sense of my notably subjective terms and requests. --Mr.98 (talk) 01:52, 19 November 2010 (UTC)

External SATA + operating system

About 8 months ago, I bought a used computer which came installed with Windows XP or Vista (I can't remember which; probably the latter). You can actually read about it here (although I ended up buying a d620 not d520). As such I have two SATA hard drives:

• One 250MB, which I use primarily and run Ubuntu on
• One ~80GB, which is collecting dust in storage

I have a few questions about this setup:

1. If I place my secondary hard drive into an external USB like this[2], will it work?
2. Is it possible to run Windows from a USB while I have the first hard disk still attached? Windows is pretty pissy about sharing a computer with another OS. And is it wise to run Windows from a USB? I'm only doing it because there are games that can't play on Linux (Wine is utterly resistant to fixing the bug plugging up Sims 2 [3], and I don't have enough knowledge of remote programming - let alone modern 3D graphics - to fix the bug for Wine).
3. Will I run into any sort of issues with the Software Protection Platform locking me out because the hardware configuration has changed (the product key was activated by the vendor before they shipped me the laptop)?

Magog the Ogre (talk) 05:28, 15 November 2010 (UTC)

For Question 1 : While yes it seems to work (All SATA drives) Are you sure it's two drives or is it just 2 separate partitions? Just check if you are not sure (Seems you don't have XP. If you do, go to Disk Management)

For Question 2 : Yes, but it may be laggy. It depends on the disk speed (most likely 7200 rpm) of the hard drive, but I don't think there would be any problems. Just make sure the hard drive is not one of those eco ones that just go to standby and takes a few seconds to spin again, or you may see lag. But that's unlikely.

No idea for Question 3, sorry. Good Luck! General Rommel (talk) 06:01, 15 November 2010 (UTC)

Well, considering that the other disk is collecting dust in the basement, yes, I'm quite sure it's not a virtual drive. ;) Magog the Ogre (talk) 06:05, 15 November 2010 (UTC)

Also some searching finally turned up [4], but I don't think I have the CD. Magog the Ogre (talk) 06:34, 15 November 2010 (UTC)

Here at work, I use Ubuntu as my primary desktop environment. The machine also came with a separate partition with Windows XP installed, and GRUB gives me the choice to boot into Ubuntu or Windows. That's fine, but I have need of more regular access to Windows XP without leaving Ubuntu (mainly because the company's Exchange server works so much better with the Microsoft Outlook mail client than, say, Thunderbird), so I have switched from the dual-boot setup to running XP in a virtual machine, and I must say it is pretty seamless. I would recommend you give it a go, and you can always delete the virtual machine if you find it's not for you.

As for potential problems with SPP, I think you would encounter some problems but you should be able to simply re-activate it (though you might have to call the activation helpdesk and explain you have shifted it to a virtual machine). 212.123.243.220 (talk) 12:30, 15 November 2010 (UTC)

Come to think of it, if you don't have an XP/Vista installation CD and you don't have the Microsoft Certificate of Authenticity still attached to the machine, you might not be able to install Windows in a virtual machine. 212.123.243.220 (talk) 12:41, 15 November 2010 (UTC)

No, I use VirtualBox, but it has no support for DirectX. Virtual OS's are not good for gaming. Magog the Ogre (talk) 12:42, 15 November 2010 (UTC)

I'm moving this section back to the bottom to get more input. Magog the Ogre (talk) 02:18, 18 November 2010 (UTC)

Samsung Syncmaster 797MB monitor OSD

What does NP, PP and NN means in following context?

81. 3k 100 hz NP

60. 0k 75 hz PP

56. 4k 70 hz NN

Jib-boom (talk) 04:44, 18 November 2010 (UTC)

81.3k is 81.3kHz the frequency of the signal the monitor receives through the VGA cable

100Hz is the monitor refresh rate

NP refers to the polarity of the input signal (eg N negative, P positive)

If you follow this link [5] you can read what it say in the manual (page 33) - though it doesn't say a lot.83.100.237.223 (talk) 05:13, 18 November 2010 (UTC)

How to reset game in old Xbox

How do I delete game memory in my Xbox so that it resets to default? —Preceding unsigned comment added by 220.255.2.65 (talk) 08:49, 18 November 2010 (UTC)

Go to memory, select the game slot and click delete.

No add/remove programs in my settings folder!! what can i do to uninstall?

it's win 2003 srv std ed. thanx --217.194.34.103 (talk) 08:56, 18 November 2010 (UTC)

Perhaps discuss it further? Has this occurred recently, has other strange things happened? Perhaps there is a virus? General Rommel (talk) 11:17, 18 November 2010 (UTC)

(e/c)

Is the icon itself missing or is the problem that when you click on the icon, the add/remove programs pane does not appear or is not populated with programs? If the first, searching suggests these possible solutions:

Make sure you are logged in as administrator

Make sure the icons are not appearing out of order (ie. the add/remove programs icon is there, but the icons are not sorted alphabetically)

Run "appwiz.cpl" (no quote marks). Does the Add/Remove programs dialog start? ([6]) - some other options there, although it is not aimed at Windows Server 2003.

If the second, consider this. Also consider rolling back any changes you have made recently. Depending on your setup, it could also be caused by a group policy setting. --Kateshortforbob _talk 11:29, 18 November 2010 (UTC)

Count valid English vocabularies

10,000+ strings consisted of a-z are generated randomly, and all of them are shorter than 5 letters. Now I want to count how many strings are valid English vocabularies, so I need a method or a program to do this work, could you help me?--Merry Rabbit 09:16, 18 November 2010 (UTC)

There are many. First you need a list of valid words - see e.g. this page. The simplest solution then is to just read the word list into an efficient data structure (a Python dict or set would do fine), and then check for each of the generated words if it is in the word list. If your data structure is O(1), the word list has n elements and your test set has m elements, overall performance is O(n)+O(m) (n for loading the dictionary with the n words from the list, m for checking each of the m words). Here is an example implementation for recent Linux/Python systems using the normal /usr/share/dict/words list as the source and a hard-coded file name for your candidates:

#!/usr/bin/env python

english = set()

# Use UNIX word list - probably not the best...
fp = open("/usr/share/dict/words")
for i in fp:
    english.add(i)
fp.close()

count = 0
english_words = 0

# Hard coded, one word per line.
fp = open("mywords")
for i in fp:
    count = count+1
    if i in english:
        english_words = english_words+1
fp.close()

print "Read ", count, " words", english_words, " are english words"

--Stephan Schulz (talk) 13:48, 18 November 2010 (UTC)

If no string appears twice in the list, you can make do without any programming. Let a be the file with your strings, one per line, and let b be a reference list of English words in the same format, such as /usr/share/dict/words suggested above. Count the number of lines in a, in b, and the number of distinct lines in their union:

$ wc -l a
10000
$ wc -l b
479625
$ sort -u a b|wc -l
487576

Then the number of English words in a is the sum of the first two numbers minus the third number (in my example, 10000 + 479625 − 487576 = 2049).—Emil J. 15:00, 18 November 2010 (UTC)

Now that I had a look on the first few pages of my /usr/share/dict/words, I wouldn't consider most of the garbage listed to be actual English words, so this is not a particularly reliable source to obtain a meaningful answer.—Emil J. 15:07, 18 November 2010 (UTC)

Encryption master key held by the Chinese

New York Times informs that the Chinese rerouted much of the Internet traffic through Chinese servers on 2 occasions last spring. The article notes that "While sensitive data such as e-mails and commercial transactions are generally encrypted before being transmitted, the Chinese government holds a copy of an encryption master key, and there was speculation that China might have used it to break the encryption on some of the misdirected Internet traffic." What key is that? What's the point to encrypt secret military emails with a code that could be broken by the Chinese government??? 93.172.63.225 (talk) 10:43, 18 November 2010 (UTC)

While "the Chinese government holds a copy of an encryption master key" is most likely true in the literal sense, it's also irrelevant. The internet is an end-to-end network, and all really sensitive data should be encrypted directly by the participants, using trusted encryption methods which do not have or require a "master key". Most protocols use asymmetric public key cryptography to transmit one-time session keys, and something like AES for the actual data. The NYT article seems to be badly confused, as it speaks about "web traffic routes" - there is no such thing. They probably mean IP routes. --Stephan Schulz (talk) 13:16, 18 November 2010 (UTC)

It's not at all clear what that means. I suppose the following might be what they mean is that, in addition to being able to route all traffic, read and change unencrypted traffic, a government-backed attacker can also compromise a certificate authority based in that country (something that's very very difficult for anyone else to achieve). That could, kinda, lead to some capacity to "intercept" SSL traffic, but at a cost. The report talks about China, which would entail compromising China's CA, and as there isn't a shred of evidence that that has happened, I'll write an explanation for the imaginary country of Brobdingnag, whose top-level name is .bro, and in which BROBNIC is the country's domain registrar and certificate authority. Here we go:

• agents of the dreaded Brobdingnag intelligence service hack into ISPs and force traffic to be redirected via a datacentre in Brobdingnag
• you visit gmail.com (with unencrypted http)
• the server in Brobdingnag sends your browser (again unencrypted) a redirect, making it go to googlemail.co.bro (a server they operate) on an https connection
• your browser tries to verify its SSL key with one of the certificate authorities it knows about a priori (e.g. Firefox's list is https://www.mozilla.org/projects/security/certs/included/) - you'll have to imagine BROBNIC is also listed there
• for that domain the certificate authority is registered with a certificate authority: BROBNIC
• as BROBNIC has been leaned on by the Brobdingnag intelligence service, it confirms the SSL certificate and says the certificate is owned by "Google Inc."
• your browser then establishes an SSL connection with that service, which acts as a man-in-the-middle proxy between you and the real google mail; they're one end of that connection, so they can read (and change) your gmail traffic.

But that article says it's all "speculation" (twice) and that there's "no evidence" of any of this actually having happened. If the above happened, lots of network admins would have noticed that the wrong CA was being asked for sites, and big sites like Google would have noticed (by dint of their own automatic testing) that stuff that should have been going to gmail.com was going somewhere else entirely. This would have immediately implicated the compromised CA. So browser manufacturers would remove BROBNIC from their list of trusted CAs. That would mean no-one would be able to establish a verified SSL connection with any .bro domain. This ruins the reputation of BROBNIC and the entire internet industry of Brobdingnag. Note this all holds for the public-key infrastructure that allows you, a regular person, to find out the SSL public key for Google using your web browser (it does entail you trusting the CAs). A more secure network (which surely any decent military operates) can rely on clients having more a priori knowledge and using clients that insist on connecting to an IP address they know with a key they already have (in practice they'd have a VPN to a specific IP with a specific key, which can't be fooled by the above mechanism). -- Finlay McWalter ☻ Talk 13:18, 18 November 2010 (UTC)

Thank you very much! Your answer is excellent and clear 93.172.57.34 (talk) 13:35, 18 November 2010 (UTC)

PPP connection with Ubuntu II

Follow-up to: [[7]] above.

Now, I am at list getting the IP and some traffic is flowing through the network.

Extended content

GNOME PPP: STDERR: --> Ignoring malformed input line: ";Do NOT edit this file by hand!" GNOME PPP: STDERR: --> WvDial: Internet dialer version 1.60 GNOME PPP: STDERR: --> Cannot get information for serial port. GNOME PPP: STDERR: --> Initializing modem. GNOME PPP: STDERR: --> Sending: ATZ GNOME PPP: STDERR: ATZ GNOME PPP: STDERR: OK GNOME PPP: STDERR: --> Sending: ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 GNOME PPP: STDERR: ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 GNOME PPP: STDERR: OK GNOME PPP: STDERR: --> Modem initialized. GNOME PPP: STDERR: --> Sending: ATM1L1DP*99# GNOME PPP: STDERR: --> Waiting for carrier. GNOME PPP: STDERR: ATM1L1DP*99# GNOME PPP: STDERR: CONNECT GNOME PPP: STDERR: --> Carrier detected. Starting PPP immediately. GNOME PPP: STDERR: --> Starting pppd at Wed Nov 17 21:59:09 2010 GNOME PPP: STDERR: --> Pid of pppd: 7875 GNOME PPP: STDERR: --> Using interface ppp0 GNOME PPP: STDERR: --> Authentication (CHAP) started GNOME PPP: STDERR: --> Authentication (CHAP) successful GNOME PPP: STDERR: --> local IP address 212.169.177.2 GNOME PPP: STDERR: --> remote IP address 10.64.64.64 GNOME PPP: STDERR: --> primary DNS address 85.62.229.133 GNOME PPP: STDERR: --> secondary DNS address 85.62.229.134 GNOME PPP: STDERR: --> Script /etc/ppp/ip-up run successful GNOME PPP: STDERR: --> Default route Ok. GNOME PPP: STDERR: --> warning, can't find address for `www.suse.de` GNOME PPP: STDERR: --> warning, address lookup does not work GNOME PPP: STDERR: --> Nameserver (DNS) failure, the connection may not work. GNOME PPP: STDERR: --> Connected... Press Ctrl-C to disconnect

But, I still cannot navigate. There's a problem (see above at the end) with the DNS and the browser cannot resolve any address.

My config file for wvdial is:

Extended content
[Dialer CARREFOURINTERNET] Modem = /dev/ttyUSB0 Baud = 115200 Init = ATZ Init2 = ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 Init3 = AT+CGDCONT=1,"IP","CARREFOURINTERNET" Phone = *99# ISDN = off Username = '' Password = '' New PPPD = yes Stupid Mode = 1 Dial Command = ATDT

I have the impression that the problem lyes in the init strings, since when I changed them, I also got an IP. By the way, what do this init strings mean anyway? Mr.K. (talk) 10:55, 18 November 2010 (UTC)

Solved, I added ,,0,0 at the end of Init3 = AT+CGDCONT=1,"IP","CARREFOURINTERNET". However, why did this ,,0,0 work? Mr.K. (talk) 13:39, 18 November 2010 (UTC)

It seems that CGDCONT thing sets up the PDP Context; this article shows the meanings of the values of its parameters, and this explains (a bit) more. -- Finlay McWalter ☻ Talk 15:36, 18 November 2010 (UTC)

Saving Hotmail emails to hard disk

What is the best way to save Hotmail emails to hard disk? Not doing them manually, as that would take weeks. I only have Outlook Express available (but not used) as I deleted MS Office some time ago. I'd prefer some freeware gismo that would just do it without too much complication. I use WinXP. Thanks 92.24.187.23 (talk) 13:36, 18 November 2010 (UTC)

You could use Windows Live Mail to export the messages.Smallman12q (talk) 15:20, 18 November 2010 (UTC)

Easy script language to back up data to a pen drive

I use WinXP. I have a small pen drive pernamently stuck into the back of my computer. I would like to be able to copy c:/mydata to e:/backup. In fact I'd like to copy to backup2, then when that is copied without any problems to delete backup1; and then next time to copy from mydata to backup1 and then delete backup2. What would be an easy script language to write this in please?

I would like this script to operate as part of the closing-down procedure of the computer: how do I get it to do that please? Thanks 92.24.187.23 (talk) 13:42, 18 November 2010 (UTC)

You'd like to make a copy of your entire hard drive each time you shut down(this would be inefficient)? Or a specific directory?Smallman12q (talk) 15:17, 18 November 2010 (UTC)

Just the small directory mydata. 92.15.5.101 (talk) 17:05, 18 November 2010 (UTC)

OK, what script languages can copy directories and their contents easily (eg "copy c:mydata e:backup"), and would be suited to the above task? Thanks 92.29.125.34 (talk) 20:43, 19 November 2010 (UTC)

A Batch file, Shell script, or a macro language such as AutoIt. Smallman12q (talk) 14:31, 20 November 2010 (UTC)

Would FreeBasic be a macro language? Thanks 92.24.181.96 (talk) 23:14, 20 November 2010 (UTC)

I think there is a windows port of rsync. 69.111.192.233 (talk) 07:48, 21 November 2010 (UTC)

Camera timer capture and filenames

I want a Windows program which can take a picture with an attached digital camera every 1 hour. So far in my searches I've only found this which works perfectly except that it overwrites the image each time it takes a new one. I tried using %date% %time% as the filename in the save as dialogue box, so that each time the program takes a picture it doesn't overwrite the previous file, but it doesn't work sine %date% and %time% output using : separate the digits, and : is not allowed in Windows filenames. Are there any other programs which I could try, or perhaps a way to get a timestamp from Windows without : in it? The OS is Windows XP 82.44.55.25 (talk) 15:28, 18 November 2010 (UTC)

What you want is a webcam program. There are literally thousands of them. You want one with archiving capability. You won't be saving the image to a webpage (the webcam part), but you will be archiving it locally. I use camE, but I use Linux, not Windows. Searching for "webcam program with archiving" will locate more programs that you will want to look through. -- kainaw™ 15:35, 18 November 2010 (UTC)

what percent of the area of a 17" macbook pro scren does a 15" macbook pro screen have?

Thanks. 84.153.193.81 (talk) 16:01, 18 November 2010 (UTC)

This sounds like homework. You know that both screens are rectangles and that the width measurement is from one corner to its opposite at a 90º angle. Do a little trig and it should be no problem to figure out their areas and thus the percentages. --Mr.98 (talk) 16:11, 18 November 2010 (UTC)

No trig required, surely ? Just use area ratio = square of linear ratio (assuming screens have the same aspect ratio). Gandalf61 (talk) 16:23, 18 November 2010 (UTC)

I wasn't sure they wanted to make that assumption. (All Macbook Pros probably have the same aspect ratio, but not all Mac laptops.) It's worth noting that the 15" is really 15.4" when you get down to it. Of course, if you wanted to take the really easy way out, you can get the pixel counts from the MacBook Pro article. --Mr.98 (talk) 16:26, 18 November 2010 (UTC)

it wasn't homework, but I have mild mental retardation and am not able to do "trig" or other math like that, I just wondered what the screen percentage is. Thanks for your understnading. 84.153.193.81 (talk) 16:42, 18 November 2010 (UTC)

If you look at MacBook_Pro#Technical_specifications, you'll see that that a 15.4 inch screen(1,440 × 900) has 73% of the pixels of a 17 inch(1,680 × 1,050).Smallman12q (talk)

...or, in round terms, about three quarters. Gandalf61 (talk) 17:02, 18 November 2010 (UTC)

Hang on if you want to compare screen size by area, are you sure both have exactly the same Pixel density? Nil Einne (talk) 07:26, 20 November 2010 (UTC)

If the screen do not have the same pixel density, and you are comparing them by screen area, then the 15.4" screen has approximately 82% of the screen-area of the 17" screen. This is, of course, presuming that both screens have the same aspect ratio. Rocketshiporion♫ 03:07, 21 November 2010 (UTC)

Flash ad archive

Is there a flash advert archive somewhere online?Smallman12q (talk) 17:00, 18 November 2010 (UTC)

Coloribus maybe? 92.15.13.70 (talk) 20:20, 18 November 2010 (UTC)

Interesting site, but it doesn't seem to have flash ads, only tv/print ads.Smallman12q (talk) 21:46, 18 November 2010 (UTC)

Restrict printer

I have a printer set up on my house PC but the kids keep printing out picture using up all my ink, is there a way I can prevent them from printing from their user accounts, their user accounts are not set up with admin privileges. Mo ainm~Talk 19:21, 18 November 2010 (UTC)

Probably, but first you need to tell us what version of Windows you are using. XP, Vista, 7? --Mr.98 (talk) 19:33, 18 November 2010 (UTC)

Low-tech solution.... Take away network access from them. No Facebook. No Twitter. No games. It won't take long for them to agree to stop printing just to get back online. -- kainaw™ 19:37, 18 November 2010 (UTC)

Tried a clip round the ear and that didn't work ;) Thought about removing network access but then they hit me with "I need it for school" so thats not an option. Using Windows XP. Mo ainm~Talk 20:59, 18 November 2010 (UTC)

Allow them to use it for school but nothing else? If that doesn't work you may have other problems, but a quick and dirty solution is just disable stuff they won't need for school like Facebook, twitter, Youtube, etc. Anyway as for your question, I would guess in most versions of Windows just click on the properties of the printer and disable all permissions for the printer for accounts you don't want to have access. Of course you need to have the security tab which may not show up in some versions of Windows. See [8] or [9] for XP specific info. Nil Einne (talk) 21:15, 18 November 2010 (UTC)

Images not showing up

One my website, Leubantia Wiki, the flags of Republic of Macedonia, Paraguay, Honduras, Romania, Croatia, Nigeria, Belgium, Estonia, Slovakia, Peru, Republic of Ireland, Gabon, Costa Rica, USA, Algeria, Georgia (country), and many, many more images are not showing up. I do not have a clue why so many images are not showing up. My computer must, but maybe not, have a virus or something... Also, when I try to re-upload them, they still don't show up. There has got to be a bug. Also, my stupid compute is making TYPOS!!!!!! —Preceding unsigned comment added by Velociraptor888 (talk • contribs) 21:46, 18 November 2010 (UTC)

I can see the flags of the Republic of Georgia and Algeria perfectly fine, as an example. I'm not about to speculate on a cause if I can't reproduce the error, and I fail to follow the logic that leads you to conclude that your computer has malware. As for the computer "making typos" -- are you sure it isn't a problem with the entity between the keyboard and the chair? Have you tried cleaning the keyboard with a screwdriver and compressed air? Xenon54 (talk) 22:02, 18 November 2010 (UTC)

IMHO, I concur with the conclusion of PEBKAC, except for the possibility of a malfunctioning keyboard or (in some applications, such as word-processing) an AutoCorrect option being turned on in the wrong language (e.g. US English instead of British English). Rocketshiporion♫ 03:13, 21 November 2010 (UTC)

November 19

TinyC for loop problem

ie Tiny C Compiler - even this doesn't work:

#include <stdio.h>
int main (int argc, char **argv)
{
  for (int a=0;a<10;a++)
     printf("%d\n",a);
}

sorry forgot the error.. identifier expected in the line with the for statement. Works on LCC, what obvious error am I making, thanks in advance.83.100.237.223 (talk) 08:40, 19 November 2010 (UTC)

I think you're typing in what is like "Borland C". I remember Microsoft Visual C being aghast at my same attempt. Anyway, I run tinyc, all I did was change your code to:

#include <stdio.h>
int main (int argc, char **argv)
{
  int a;
  for (a=0;a<10;a++)
     printf("%d\n",a);
}

and it works fine... 84.153.193.81 (talk) 10:18, 19 November 2010 (UTC) And

It's only since C99 that the first expression of the for-statement may take the form of a declaration. According to our article, the Tiny C Compiler is a C89/90 compiler with some C99 features. decltype (talk) 10:28, 19 November 2010 (UTC)

Thanks, that's it.77.86.15.94 (talk) 17:40, 19 November 2010 (UTC)

Resolved

Cloud computing question

tell me about cloud computing and software as a service(saas) model. —Preceding unsigned comment added by Pawantechdell (talk • contribs) 10:38, 19 November 2010 (UTC)

Cloud computing and Software as a service. Once you've read those, if you have any specific questions feel free to come and ask. --jjron (talk) 13:19, 19 November 2010 (UTC)

Welcome to Wikipedia! In addition Cloud computing and Software as a service, you may wish to read our articles on Cloud applications, Cloud clients and Cloud APIs. Rocketshiporion♫ 13:35, 19 November 2010 (UTC)

BBC News website stats

Hello all, I'm trying to find some reasonably accurate stats for visits and unique visitors to the BBC News website. I've had a look at Alexa, but it only offers figures for the entire BBC site (www.bbc.co.uk), and in any case, these figures appear to be listed in terms of percentage-reach rather than actual figures. I have a few free toolbar apps from SeoBook installed, but they also only offer stats for the entire site. I was wondering if any of you super-helpful Wikipedia Computer Helpdesk folk know of anywhere I could get some figures from.
To Summarise

• I'm looking for the number of page views of the bbc.co.uk/news directory (A recent as possible, and doesn't have to be 100% accurate, but would be good if they were reasonably indicative)
• And the number of unique visitors

Any help, or any direction to useful tools or apps is much appreciated. Thanks all, Darigan (talk) 11:07, 19 November 2010 (UTC)

Screen resolution with two dissimilar monitors changes on reboot

My working environment at work consists of a Dell Optiplex 755, running Ubuntu 10.04, which drives two LCD screens through a DVI splitter cable. The primary screen is a Dell 24" monitor connected to #2 on the splitter cable, and the secondary screen is a Dell 17" monitor connected to #1 on the splitter cable. I use the big monitor for software development and the small monitor to display a virtual machine running Windows XP. I occasionally have to reboot (usually when required by a Ubuntu update) and that's where the trouble begins. After rebooting, both monitors display in 1024x768 resolution. The last few times, it has taken several reboots and much messing about with Ubuntu's monitors preference panel to get the big monitor to display in 1920x1200 (and have circular objects shown as circles instead of ellipses) and the small monitor to display in 1280x1024. Is there a way I can get Ubuntu to recognise my monitors' correct resolution after a reboot, or at least have it remember my preferences? Astronaut (talk) 12:35, 19 November 2010 (UTC)

Edit xorg.conf. I don't know the details, but google is your friend. --85.77.53.134 (talk) 14:46, 19 November 2010 (UTC)

Does Ubuntu still use xorg.conf? -- kainaw™ 16:26, 19 November 2010 (UTC)

I was about to say yes, but it looks like its gone now. Paul (Stansifer) 20:42, 19 November 2010 (UTC)

Xinput has replaced xorg.conf for input device configuration a long time ago, but I think you can still use it for monitor settings. I'm not sure if the file exists out of the box. --85.76.179.108 (talk) 08:11, 20 November 2010 (UTC)

Is the problem that the preferences you set in Display Preferences aren't "sticking"? Take a look at the file ~/.config/monitors.xml. Is it correct now? When the displays go back to the wrong settings, check to see if it changed also. Paul (Stansifer) 20:42, 19 November 2010 (UTC)

TBH, I can't tell if the preferences are not sticking or if the computer is trying to "read" the first monitor it sees and guess the required resolution from that (the smaller monitor wakes up several seconds before the big monitor). Another weird thing is the big monitor is connected to #1 on the splitter cable; if I swap them around, the menu & icons end up on the small screen. I would have thought the "main" monitor would be on #1. Unfortunately, I'm away from my desk right now. I'll be able to check if I have xorg.conf, Xinput or ~/.config/monitors.xml on Monday. Astronaut (talk) 00:36, 21 November 2010 (UTC)

Saving all glipper (or clipboard) items

I have Glipper installed (in Ubuntu), but do not know how to save all the items at the same time in it. Apparently the program doesn't have a function for that, only allowing picking individual items from a list.

Does someone know how can I save all glipper (or clipboard) items? Copy-and-pasted items must be in some file somewhere, but I don't know where...

I would also be happy about alternatives to save a series of copied items into a file (without going through the list one by one, of course). Mr.K. (talk) 17:25, 19 November 2010 (UTC)

Facebook apps that tell you who visited your profile?

That's a rumor in my eyes. Would anyone please confirm or deny whether there is such an application that lets you see who has visited your profile? (An account, or an IP if someone was logged off while visiting it.)

If there is such an app, please link. Thanks. --129.130.252.148 (talk) 18:45, 19 November 2010 (UTC)

It can't possibly work. Facebook's privacy policy shields this sort of information. ╟─TreasuryTag►Africa, Asia and the UN─╢ 18:46, 19 November 2010 (UTC)

Someone said in one of the app games I play that the one player they were talking about could see me visit their profile. I asked, "You mean he has an app that logs visitors and tells him who viewed his profile?" Their response was, "Who doesn't?" How are you so sure that such apps can't possibly work? Why couldn't those apps circumvent the "privacy policy shields?" --129.130.252.148 (talk) 19:18, 19 November 2010 (UTC)

How are you so sure that such apps can't possibly work? You asked the question. I gave the answer. If you don't believe me, that's fine. If you think that Facebook would accept applications actively violating its terms of service, feel free. But you'd be wrong. ╟─TreasuryTag►UK EYES ONLY─╢ 19:20, 19 November 2010 (UTC)

Just because it shouldn't be able to be done doesn't mean it can't be done. Over the years there have been loads of facebook hacks ranging from seeing private photos to updating peoples profile status without logging into their account. Facebook usually patches these fairly quickly, but no doubt there are more undiscovered hacks perhaps being exploited by people, privacy policy or not. 58.9.2.225 (talk) 19:28, 19 November 2010 (UTC)

It's true that Facebook has tended to be very 'leaky' and has had many flaws in their privacy and security code (I don't know if I'd call many of the things 'hacks' per se). However I can't find any evidence of any leak on who visited your profile. hubpages.com/hub/How-To-Get-A-Facebook-Tracker (blacklisted site) shows someone who looked over a year ago with no luck. [10] mentions there are many scam apps which claim to do so which may be the source of such rumours. Nil Einne (talk) 20:24, 19 November 2010 (UTC)

One such app I noticed about half a year ago was something like 'which friends are viewing your profile the most'. A few of my 'friends' activated this and I was, let's say, interested to see in my feed that I was listed in the top 20 out of about 400 for a friend whose profile I'd only viewed about once or twice previously, probably on the day she friended me about 8 months previously, and maybe a second time when she changed her profile pic. So either she's not getting many friends viewing her profile or it was simply randomising a top 20 list from all her friends. I'd never even commented on anything on her profile, so it wasn't even that sophisticated in generating the list. This randomising theory was confirmed when a second friend ran it twice within about an hour, producing two entirely different lists. --jjron (talk) 02:50, 20 November 2010 (UTC)

Yahoo! Chat changed years ago. Why?

Sometime in 2005 or so, I could no longer access chat the way I used to. Why did they change it so substantially? What can you tell me about the history and circumstances behind the change of Yahoo! Chat? --129.130.252.148 (talk) 18:45, 19 November 2010 (UTC)

In 2005, Yahoo started making a lot of changes so it could work with other services and be accessed through a web interface. The main idea was interoperability. While AOL was making it difficult for others to work with their chat, Yahoo decided to invite interoperability. -- kainaw™ 19:12, 19 November 2010 (UTC)

<personal rant> IMO Yahoo! have progressively worsened their service over the years, whether by design or accident I don't know. For example for ten years I was able to have the apostrophe in my surname in my profile, a couple of years ago they 'updated' their service, and the apostrophe now displays as an ASCII code markup instead, meaning I have to spell my name incorrectly (sans apostrophe) in order to keep using their service. This is pretty unacceptable in this day and age. When I complained they made some rumblings about looking into it, but never did anything - surely this should be a simple fix? Then about a year ago when trying to introduce their social networking feature, they basically said they were going to make email address books public, listing all your contacts, unless users opted out (talk about aping Facebook). I of course opted out, and I've attempted to basically shut down all 'social networking' features on Yahoo! related to their email, including Chat, as I'm not at all comfortable with their security any more. Maybe this drop off in personal security is related to the interoperability Kainaw talks about, but if anyone from Yahoo! is reading this, please lift your game - you used to be a lot better. --jjron (talk) 03:06, 20 November 2010 (UTC) </personal rant>

I don't work at at Yahoo! and never have... but interestingly, the company I am currently working for moved into the building Yahoo! vacated, during the massive Yahoo restructuring after the dot-com bubble-burst (around 2003- or 2005, when their software also started suffering badly). We still have Yahoo!-colored purple walls and break-rooms. The decor and furniture is all purple and yellow and green. I have also worked in one of the original Fairchild Semiconductor buildings - (not the Shockley Semiconductor Laboratory, but the one they moved to a few years afterward). If there's a point here, I guess, its that these vacated buildings are artifacts of companies that have passed their prime, and been replaced. It happens to the best... Nimur (talk) 05:48, 20 November 2010 (UTC)

forget SQL language - what IS, mathematically, a relational database?

Forget sql. Could you briefly SHOW me what a relational database actually IS, mathematically, by writing pseudocode implementing any part of it. It can be very high-level pseudocode, English if you want, and you can refer to any mathematical concept and I will go look it up and learn it if I don't know it already. Thank you SO much. (this is not homework). 92.230.70.89 (talk) 20:07, 19 November 2010 (UTC)

First some links Relational database gives an overview, but Relational model is the better article for a more mathematical or abstract description.

To get a feel for it I recommend looking briefly at other database model types - eg Hierarchical database model and Network model (database), after which it should be clear (if it wasn't already) what the 'relational' part of 'relational database' refers to.

To a attempt a brief description - a relational database is a computer implementation of a table (not necessarily 2 dimensional - could be a single column, or have more dimensions) - to cap it and make it a database the idea of a key is added ie an ID (such as an account number or National identification number) which is used to link items between sets of tables. (typically a relational database will consist or more than one table)

In computer science terms each table is basically some sort of array.77.86.15.94 (talk) 20:48, 19 November 2010 (UTC)

The article Database model also looks helpful - the images look as if they will be self explanatory.77.86.15.94 (talk) 20:59, 19 November 2010 (UTC)

Personally I think a mathematical understanding is not really the best way to understand what a relational database is. It's not an equation or a function. It's a means of organizing and cross-referencing information. If you're interested, I'd be happy to elaborate. I find the best way to explain them is through analogies to "real life" datasets. For example, if your Rolodex was cross-referenced with the list of people who you call every day, and that was cross-referenced with a description of all of the subjects you like to talk about. The relational database lets you hold all of that information in the same place and have it operate together seamlessly. It's really just about as simple as that — multiple tables of data that have defined relationships for how the data co-exists. Math or even code are not the best way to understand the database structure itself; the code just lets you manipulate the data or ask questions of the database, but it doesn't really define the structure of the database, which is more of an abstract entity that sits above the code you actually use to access it most of the time. --Mr.98 (talk) 22:43, 19 November 2010 (UTC)

Are you familiar with C/Java-like programming languages? A table is similar to a C struct type (or a class with public data members), and each row is similar to an object of that type (except that tables know their own rows, while classes don't usually keep track of their instances). A key is essentially a pointer, a foreign key is a field of one struct that points to another struct, and referential integrity means that there are no dangling pointers.

The programming language Prolog works a lot like a relational database, and you might find it easier to learn and experiment with than your typical RDBMS. -- BenRG (talk) 23:29, 19 November 2010 (UTC)

...and forgetting relational databases, what is screaming fast?

If we move away from relational databases. What kind of database, then, has a reputation for being screaming fast? 92.230.70.89 (talk) 20:27, 19 November 2010 (UTC)

I'm assuming you didn't mean what brand, or what type of hardware implementation.. Although I can't answer this properly it's worth pointing out that most non-trivial database querys will involve iterations to find matches - and this is improved by additional indexing (as a first resort, other stuff helps) - the indexing includes such things as storing entry points into tables of data as sorted by various values (eg in a table of people by name, address, town, fav. color etc index points for entry to the table sorted by each field can help, along with linked-list type data structures that connect similar fields - eg a linked list for each fav. color that connects each sub-set of people)

The overall effect of such optimisations is to overlay on top of the original relational database additional supplimentary database structures that are conceptionally pretty much the same as hierarchical or network based databases.. similarily you can probably think of examples were adding an indexing overlay of a relational type could help searches within a hierarchically organised database

So the best type is a mix - what I've described above are simple optimisations - I see that I haven't really answered your question at all, but seeing as you asked above about implementations of databases I though this might be useful (There's a brief introduction to when to index here [11]. There's always more ways to optimise.

In case you didn't know a good search term for this (the equivalent of Mips Flops or Mhz in processors) is "Database transactions per second" like Ghz the figures are not necessarily meaningful in the real world. Apologies if all of that was irrelevant to your needs.77.86.15.94 (talk) 21:45, 19 November 2010 (UTC)

It kind of depends on what you want to do with it, is not a very helpful but a somewhat more optimal answer. If you are really concerned about database performance, the trick is to figure out how to divide up the data in clever ways that optimizes for the kind of usage you're going to use. That will vary depending on the data in question. This is a separate issue from hardware and software performance, which are largely separate from the question of structure. --Mr.98 (talk) 22:45, 19 November 2010 (UTC)

Screaming fast is the kind if fast that makes you scream and would cause instant death should it stop suddenly... like a reaaaalllly fast rollercoaster or a sportscar driven in a manner likely to get you an instant driving ban, or that plummet towards the ground attached to a bungee rope. However, in terms of databases it is really just marketing speak for "we have made some effort to make our database seem faster than some of our competitors". Ignore such hyperbole unless you are in the business of selling databse products. Astronaut (talk) 01:09, 21 November 2010 (UTC)

Generally, this is very much dependent on the task. If we go with searching (as there are a number of factors the affect speed of updating and inserting data), the primary question is what you will be searching on. If your intent is to search on a single index, and you need a lot of data, then flat file databases are extremely good. The search algorithm for an ordered list is quite fast, and a flat file database doesn't need to make the joins that would be required in a relational database. However, if searching for a non-indexed term in a large flat file database, the format can become very slow. The smaller data sets in a table in a relational database can have the advantage here, especially if you don't need to make a join. Optimising relational databases for speed involves, amongst other tasks, the reduction of the number of joins required for common queries (noting that this can increase the size of the database as a side effect), so if you search doesn't involve joins you can expect a nice speed increase. - Bilby (talk) 01:24, 21 November 2010 (UTC)

November 20

Uses of Microsoft Silverlight

I just downloaded and installed Office 2010 Professional (don't worry, as an IU student, I'm allowed to download it free from http://iuware.iu.edu), and it came with Microsoft Silverlight. Not being a tech savvy person — I can't use Excel or Access beyond putting data into the cells, and I'm hopeless with Powerpoint — I don't foresee a use for this program. Any ideas if I'm likely to encounter situations in which I'd find this program useful, or would it be more helpful for me simply to uninstall it? Nyttend (talk) 02:59, 20 November 2010 (UTC)

I recently had to install Silverlight to view a particular graphics rich website. Basically I think this program is much more about viewing certain graphics rich websites (much like Adobe Flash) than using your standard Office applications. In other words, I'd suggest to leave it. I haven't heard of it doing any harm. --jjron (talk) 03:12, 20 November 2010 (UTC)

It is (as above) pretty much a competitor to flash (as used on Youtube, and many other places on the web) - in my experience its use is very rare - there's some examples here. http://www.silverlight.net/showcase/ also found this http://memorabilia.hardrock.com/ via a forum.

In general its use is very rare, you can almost certainly uninstall without noticing the difference if the disk space is valuable to you.77.86.15.94 (talk) 03:49, 20 November 2010 (UTC)

Notably, Netflix uses Microsoft Silverlight to stream subscription videos using digital rights management. Nimur (talk) 05:54, 20 November 2010 (UTC)

One of the most conniving puzzle games on the net uses it (highly addictive, click through at your own risk!): http://robozzle.com. Also I've seen online mind mapping software use Silverlight (kind of like a Google Doc or Office, editing live on the net). As mentioned, it is a competitor to Flash, might as well leave it installed, you'll need it one day for some website anyway. Zunaid 07:21, 20 November 2010 (UTC)

MSN utilizes silverlight as do certain other Microsoft subsidiaries/partners.Smallman12q (talk) 14:26, 20 November 2010 (UTC)

Faceboook - suggesting people you may know

Hi How does Facebook suggest people you may know? When a person displays my profile, does it mean that later Facebook will suggest me adding this person to my friends? —Preceding unsigned comment added by 178.181.131.108 (talk) 13:44, 20 November 2010 (UTC)

We've speculated on this before and we aren't sure. Some of the suggestions are clearly friend-of-a-friend or "went to the same institution/work/town" as you. But some seem to be really off the wall and unrelated to that. I don't think it has to do with you visiting anyone's page. But we don't know the "secret sauce" behind the algorithm. --Mr.98 (talk) 14:06, 20 November 2010 (UTC)

watch

how do i watch this

<rapidshare .avi.00# links culled> --Mr.98 (talk)

You download all of the component files. Then you have to combine them. Google ".001 joiner" or "file splitter/joiner" and you'll find lots of programs that can do this. Then you open the resulting .AVI file with a movie viewer. --Mr.98 (talk) 15:09, 20 November 2010 (UTC)

An easy to to do this is with JDownloader. Enter the links into it and it will download and extract them all for you. On rapidshare free service this could take several hours, so Jdownloader is so much better than doing it manually. 82.44.55.25 (talk) 15:50, 20 November 2010 (UTC)

Windows - No Disk - Exception Processing Message

This message pops up on the screen at the end of the Windows boot process (after it is populated with icons);

Windows - No Disk Exception Processing Message c0000013 Parameters 75b6bf9c 4 75b6bf9c 75b6bf9c

There are buttons to Cancel, Try Again, and Continue, but clicking on them is rejected with a loud buzzer. The same with the X delete at the upper right of the information box.

Ordinary applications at first glance seem to work okay. System functions: I was able to free up disk space but defrag and backup are not responsive.

Any ideas what's happening and what I might do? --Halcatalyst (talk) 16:59, 20 November 2010 (UTC)

Clicks Agent

Clicks Agent is a company said to have been developed by a teenager in his bedroom and sold for $40million. 1) Am I correct in thinking that its business model is that you pay it some money, and it gets traffic to your website? 2) Why was someone willing to pay $40M for it rather than replicating it themselves? I mean, it was created by a teenager in their bedroom so it can't be too difficult. 3) Where does the traffic come from? Thanks 92.15.27.119 (talk) 17:56, 20 November 2010 (UTC)

1) yes. 2) it was paid in the form of stock, so in reality was less than that. Replicating the technical side might be easy, but you'll also have a portfolio of clients, portfolio of contractors, a brand name. 3) I guess from pop-down, pop-ups, parked urls, and so on. Mr.K. (talk) 21:46, 20 November 2010 (UTC)

laptop cannot see dvd writer

Hi I have a Samsung R510 and I've had to remove the default Vista and install Windows 7 64-bit on it. Now everything works except the dvd writer... it cannot be seen in devices and cannot be seen by Nero 10. Samsung official website does not have drivers for it. Any ideas please? Sandman30s (talk) 18:36, 20 November 2010 (UTC)

Try this tool: [12].--Best Dog Ever (talk) 06:50, 21 November 2010 (UTC)

Python + Ubuntu + controlling a program

I know that you can start a program or execute a command within Python with (stream = os.popen('myprogram or command with parameters'). The result of that is a wrapper object stream. However, if I want to interact with the program, (for example, get information from this wrapper object), how do I proceed? Mr.K. (talk) 21:30, 20 November 2010 (UTC)

I believe that os.popen has been deprecated in favor of subprocess.Popen (which is a class). Once you have a Popen object, you can use communicate() on it. Paul (Stansifer) 22:22, 20 November 2010 (UTC)

OK. Thanks so far.

I am trying:

import subprocess 
stream = subprocess.Popen("klipper", stdin=subprocess.PIPE, stdout=subprocess.PIPE)
stream.communicate()
???

And then? How can I interact with a concrete program, (Klipper in this case). Mr.K. (talk) 23:36, 20 November 2010 (UTC)

popen is for processes that run in a terminal window and interact via stdin and stdout. It won't be much use for a GUI app like Klipper. You could try using the dbus package for that. -- BenRG (talk) 00:04, 21 November 2010 (UTC)

Dbus is probably too complicated for me (I only need it for a while). However, I can substitute Klipper for a command-line tool: xclip.

import subprocess 
stream = subprocess.Popen("xclip", stdin=subprocess.PIPE, stdout=subprocess.PIPE)
stream.communicate()
???

Normally, xclip outputs what I have in the clipboard. But stream.communicate() does not output this information... :( Mr.K. (talk) 01:40, 21 November 2010 (UTC)

How to move a macro in Microsoft Word

This seems like a very silly question, but I cannot make it work. I have a macro that apparently is saved in Normal.dotm. It is accessible from every Microsoft Word document I create. However, I only need it in one particular document, Document.docx, say. Is it possible to move the macro from Normal.dotm to Document.docx, so that the yellow warning message doesn't appear for every Microsoft Word document I create/open?

I have tried to open "Mallhanteraren", as it is called in my Swedish copy of Microsoft Word 2010. ("Template Manager" is a direct translation. This dialog box can be opened using a push button in the Macros dialog box.) In "Template Manager", under the tab sheet called "Macros", there are two list boxes, called "In Document.docx" and "In Normal.dotm". Between these list boxes, there are three buttons: "Copy", "Remove", and "Change name...". One would expect to be able to use this dialog to copy the macro from Normal.dotm to Document.docx, but the problem is that the Normal.dotm list box is empty, even though Microsoft Word itself tells me that my macro is indeed stored in this very document/template file (when looking in the "Macros" dialog box, for instance)!

What can I do? --Andreas Rejbrand (talk) 21:40, 20 November 2010 (UTC)

November 21

Full Disk Encryption: RAID Controller & Self-Encrypting HDDs

Hello Everyone,

   This is not a homework question.

   The objective is to create an RAID array with hardware-based encryption. Will either or both of the following configurations work?

1. Encryption-capable RAID Controller, with Self-Encrypting HDDs.
2. Encryption-capable RAID Controller, with regular (i.e. non-self-encrypting) HDDs.

   Thanks in advance to all RefDesk volunteers. Rocketshiporion♫ 03:35, 21 November 2010 (UTC)

Both will work. Each has potential security holes but will be much more secure than using no encryption. In particular, hardware disk encryption usually uses ECB mode. See that article for related issues. 69.111.192.233 (talk) 07:25, 21 November 2010 (UTC)

Core i5 vs i7

I have an option of a laptop with an Intel Core i5 460M processor and one with an Intel Intel Core i7-720QM Processor.

• Core i5 460M- 2.53 GHZ Turbo Boost up to 2.8GHZ
• Core i7-720QM- 1.6 GHz Turbo Boost up to 2.8GHz

The normal speed of the i7 (without Turbo Boost enabled), i.e. 1.60GHz, is far less than that of the i5, so why is the i7 more expensive and supposedly more faster? Thanks. --119.155.8.225 (talk) 05:12, 21 November 2010 (UTC)

The i5-4xxM has two processing cores while the i7-xxxQM has four processing cores. Dragons flight (talk) 05:22, 21 November 2010 (UTC)

So the i7 has a max speed of 11.2 GHz while the i5 has a max speed of 5.6 GHz (half that of the i7)? --119.155.8.225 (talk) 05:59, 21 November 2010 (UTC)

cell phone as workstation

Current fancy cell phones are comparable in power and capacity to desktop computers of maybe 5 years ago (1 ghz processor, 1gb ram, 32gb flash "disk"), and some run various Linux-based OS's. Some of them have HDMI output also. Does it sound practical to set one up as a general purpose Unix workstation, using a bluetooth keyboard and mouse, and an HDMI monitor? Thanks. 69.111.192.233 (talk) 06:19, 21 November 2010 (UTC)

What exactly would you use it for? Also, I'm not aware of any cell phones with 1 GB of memory. I know of some that have 512 MB of RAM. It's confusing because manufacturers in the cell-phone industry call secondary storage "memory." You'd be able to use a cell phone like that for checking your e-mail, word processing, and browsing the web. But it wouldn't be ideal for 3D, gaming, or editing video.--Best Dog Ever (talk) 06:48, 21 November 2010 (UTC)