Wikipedia:Reference desk/Archives/Computing/2010 August 25

= August 25 =

Installed GIMP 2.6.10 and GhostScript - still can't open EPS
I've installed GIMP and GhostScript on a Windows Vista computer. When I want to open an EPS file it sends this message

Encapsulated PostScript image Message

Error starting Ghostscript. Make sure that Ghostscript is installed and - if necessary - use the environment variable GS_PROG to tell GIMP about its location. (Failed to execute child process (No such file or directory))

GIMP Message Opening 'C:\Users\Admin\Desktop\pictures\aa1.eps' failed:

Could not interpret Postscript file 'C:\Users\Admin\Desktop\pictures\aa1.eps' Any ideas on how to fix this? The EPS file isn't corrupted since I was able to open it using Adobe PhotoShop in another computer. -- Lenticel ( talk ) 00:32, 25 August 2010 (UTC)


 * Have you tried setting the environment variable, GS_PROG, to point to the GhostScript installation directory? (Here's a quick how-to if you don't know how to set this).  This forum post on the official Gimp website says that some users have found it easier to copy GhostScript into the GIMP install directory - this seems heavy-handed, but will probably work.  Nimur (talk) 00:58, 25 August 2010 (UTC)
 * I've copied the files in ghostscript's bin and lib folders directly to GIMP's and it worked! Thanks :)-- Lenticel ( talk ) 01:39, 25 August 2010 (UTC)

Sending a big file
I live in NJ and I want to send an audio file (lecture) to someone in Oregon but it's 26.4MB and too big to email. After compression, it's down to 26.2MB -- still to big. Must I copy it to a CD and snail mail it? What are my other options, pray tell?  DRosenbach  ( Talk 01:37, 25 August 2010 (UTC)
 * Rapidshare, Megaupload, Mediafire, wikifortio, etc, see File sharing services. You could also rar or 7zip the file into smaller chunks which would be small enough to email, then reassemble at the other end 82.44.54.25 (talk) 02:23, 25 August 2010 (UTC)
 * Thanx! DRosenbach  ( Talk 02:57, 25 August 2010 (UTC)

Disk recovery
Hi. I have data on a disk which I want to recover; linux suse 11.1

The disk has been damaged in some way. The upshot is this: I can boot up my computer using the damaged disk; but then the machine suffers from numerous problems including inability to mount anything such as a USB stick or another hard disk, or indeed an external floppy disk. So I can't copy information out. I can boot up my computer using another disk (suse 11.1/2/3) but then the system will not mount the original damaged disk.

The only files I really need are half a dozen text files. Does anyone know a way to extract information from a drive without mounting it? 131.111.23.212 (talk) 12:18, 25 August 2010 (UTC). Ooops wasn't logged in. Robinh (talk) 12:20, 25 August 2010 (UTC)


 * Have you tried booting from a LiveCD and trying to mount the damaged disk from there? -- Finlay McWalter ☻ Talk 12:42, 25 August 2010 (UTC)
 * Hi Finlay. Yup, tried that (using the suse 11.1/2/3 installation disks).  No joy. The weird thing is, that I can *see* the files when I boot up from the damaged disk.  Cheers, Robinh (talk) 12:44, 25 August 2010 (UTC)#
 * Oops, sorry, I misread your question a bit. So you can read the files, but you can't attach another disk to safely preserve them. Does the bad boot still do the network okay? If it does, you can copy it off to another machine with SCP/SFTP. If you're desperate, and this machine (and another you have access to) has a working serial port, you can copy files over RS232 with a null-modem cable (if it's only a few files, you can do this one-at-a-time, manually, with no additional software). I know the system is in trouble, but have you tried burning a CD/DVD? -- Finlay McWalter ☻ Talk 12:59, 25 August 2010 (UTC)
 * Hi again Finlay. Thanks for this.  Can't burn a CD/DVD either.  I don't understand what you mean about RS232 and a null-modem cable.  What do I plug into the other end of the serial port?  Otherwise, Emil's suggestion might be all I have...Robinh (talk) 13:04, 25 August 2010 (UTC)
 * A null modem cable is a serial cable (almost always the 9-pin kind) with a PC at either end. For fancy operation you either run TCP/IP over that connection or a fancy file transfer program, but in your emergency case you'd cat a file into the appropriate serial device at one end, and at the other cat the serial port into a file. -- Finlay McWalter ☻ Talk 13:09, 25 August 2010 (UTC)
 * The bad boot machine still boots, but as of just now it doesn't recognize the keyboard when in GUI mode (tried three keyboards) although it *does* recognize the keyboard when I'm asked for a screen resolution at bootup. Which kinda makes life difficult.  I thought that control-alt-f1 was supposed to open a virtual terminal, but it doesn't (probably because the keyboard doesn't work)-:.  Anyone got a way round this? Robinh (talk) 14:16, 25 August 2010 (UTC)
 * If the keyboard works at bootup, chances are that it will work in text mode. Did you try to boot into run-level 3 (or whatever number is Suse using for non-GUI mode)?—Emil J. 14:24, 25 August 2010 (UTC)
 * Ahha, that sounds a good idea. How do I do that? Robinh (talk) 14:25, 25 August 2010 (UTC)
 * Depends on the boot loader. If you use grub: hit e in the boot menu, this will give you an editable list of parameters passed to the kernel. Append 3 (or, if there already is a lone number on the line, replace it with 3), hit enter. (Here's how it may look like: .) Now that I think about it, given the screwed up state of the system, it might be even better to use the single user mode (which avoids starting up the usual demons, which are unlikely to work properly anyway): the kernel parameter for that is named single (instead of a number).—Emil J. 14:39, 25 August 2010 (UTC)
 * If it's really just a handful of text files, you can use more, a camera, and OCR.—Emil J. 12:50, 25 August 2010 (UTC)


 * A really nasty way, if you can still access the boot loader, is exchanging the init=... command on the line with the kernel parameters (or placing one there if there is none) with init=/bin/bash. That will drop you directly into a shell. Note: Most likely, your hard disk will be mounted read-only when you do that.
 * You could configure an ip address manually by using

ifconfig eth0 a.b.c.d
 * (where a.b.c.d is an unused address from your home network, which usually starts with 192.168.).
 * After that, you could copy the files to a different machione using scp/sftp, as suggested above.
 * Another method, assuming you have netcat installed on both the flaky computer as well as the one you wish to copy the files to, would be this:

On the flaky machine: cd /directory_where_files_are_located tar -cvf - your file names here separated with blanks|nc -l -p 4223 -q 2

On the target machine: cd /directory_where_you_want_to_save_the_files nc -q 2 a.b.c.d 4223|tar -xvf -
 * Note: Some distributions call nc by its full name, netcat. -- 78.43.71.155 (talk) 18:27, 25 August 2010 (UTC)
 * Thanks guys (OP here). The machine is locked up right now in the IT office but I'll have a bash (literally, thanks 155!) tomorrow.  Best wishes and thanks again, Robinh (talk) 19:04, 25 August 2010 (UTC)
 * Addendum: At the bottom of http://aplawrence.com/SCOFAQ/FAQ_scotec7getnetcat.html there's a netcat implemented in Perl, so even if there's no netcat installed on the flaky machine, you could squeeze all that into one line like

cd /directory_where_files_are_located && tar -cvf - your file names here separated with blanks| perl -e"use IO::Socket;$host=shift @ARGV;$port=shift @ARGV;$socket=IO::Socket::INET->new(PeerAddr=> $host, PeerPort=> $port, Proto=> 'tcp',Type=> SOCK_STREAM) or die 'Can't talk to '.$host.' at '.$port;while (<>) {print $socket $_;} close $socket;" ip.of.target.machine 4223


 * Note that if you have to resort to this method, you have to start the target machine first and use the following commands there:

cd /directory_where_you_want_to_save_the_files nc -q 2 -l -p 4223|tar -xvf -
 * Also, this might not terminate automatically, so after it shows the last file name, you might have to press Ctrl-C to get back to the shell if it doesn't appear after 2 seconds (that's what the -q 2 is for). -- 78.43.71.155 (talk) 21:09, 25 August 2010 (UTC)


 * Very interesting stuff here. I think I'm going to bookmark the archive version of this conversation.
 * Here's my contribution: If you can't make the above stuff work, just pull the disk out physically and put it in an enclosure.  You can get 'em at Fry's for twenty bucks or so.  Then you can just attach the enclosure to the USB port of another machine and mount it.
 * It's easy; no soldering required or anything. (Except, if the bad machine is a laptop, getting the drive out can be a challenge, depending on the model.)  It's not completely risk-free though &mdash; with a hard drive that's on its last legs, there's a chance that all that manipulation could send it to computer heaven. --Trovatore (talk) 19:14, 25 August 2010 (UTC)

Malicous Virus needed
Where can I download a truly terrible virus? I need to find a virus that would extremly screw up any computer. A virus that would mess around with a computers important files and preferably a virus that would engrave some disturbing or pornographic images on the computer, that won't come off. Also it would be prefered if the virus could be easily reached and downloaded. A few months ago one of my friends told me about a virus he got that deleted his operating system and permantly engraved a picture of a old man jerking off in the background that could'nt be taken off. A virus of that magnitude would be great! Thanks in advance, and I would appreciate any help. Wikiholicforever (talk) 13:53, 25 August 2010 (UTC)


 * If you really want to screw up your computer, consult you local hardware store, they will offer a selection of fairly terminal viruses.—Emil J. 14:18, 25 August 2010 (UTC)


 * That sounds fictional, or at least highly exaggerated. There is no realistic way of permanently "engraving" an image into a computer.


 * Anyway, I'm not sure that there is a legitimate source of computer viruses that'll just hand them out to random people off the street. You're left with various shady sites that prey on would-be hackers, or contacting the right folk in chat-rooms and whatnot. If you know how to code, I suppose one of the infamous Virus Construction Kits is what you're after. They might be slightly easier to find, but probably require some skills to make them work. Wikipedia is not really the right place to ask. Asking here is like walking into a Library and asking the librarian where you can buy some heroin.
 * (Remember of course, that destroying someone's computer is a criminal offense. If they (or their parents!) decide to report it, you'll be locked up.) APL (talk) 14:48, 25 August 2010 (UTC)


 * For a kiddie, Smeg would be too complex. Best to start off with something like Virus Creation Laboratory. Either way, you'd be trespassing and committing computer crime.Smallman12q (talk) 15:28, 25 August 2010 (UTC)

I assure you all, my intentions are not criminal or intending to damage peoples property. I just want to do this for research, to see how badly I can screw my computer up before I throw it away (I thought this is what wikipedia was all about, LEARNING). I am not looking to create a virus, I just need to find a really terrible one (it may not be as bad as I described but still pretty dreadful), I know its hard to find one but thats why I came here. Please don't delete my question, as it is only for research purposes only. A link to a download would be just enough for me or a link to a site full of viruses. Thanks again and sorry for the misunderstanding. Wikiholicforever (talk) 17:02, 25 August 2010 (UTC)
 * Of course. 'Research'.   You might check out Astalavista. They've got all kinds of shady stuff. I'm not sure if that includes viruses, but IIRC they've got forums and stuff, so if you're absolutely certain of your own computer's security arraignments, you could go ask there.  Like I said before, you're not going to find a non-shady source of viruses unless you can show credentials as a legitimate researcher, which I'm sure you can't.  APL (talk) 14:57, 26 August 2010 (UTC)

File Sharing b/tween Vista and Win 7 - Problem
I got Win 7 a while back, and was able to set it up so that my Vista machine could access the computer with full read/write access, and vice-versa. There was no problem at all, and I was actually pleasantly surprised with how easy it had been compared to earlier versions. However, now, since about a week ago, my Vista laptop is giving me a message saying I do not have permission to access the Win 7 laptop - it won't even give me the 'enter your password' bit. I have no idea how or why this suddenly started to happen, but, in any case, I am now finding it difficult to set up a connection again (it was so easy last time that it just wasn't memorable). Can anyone help me either fix problem? The Win 7 laptop can still access the Vista one with no problem at all. -- KägeTorä - (影虎) (TALK) 15:25, 25 August 2010 (UTC)
 * Difficult to troubleshoot without access to the computers. I had a problem with my XP laptop all of a sudden giving the same message with my guest account setup on win7. Then I went to control panel, user accounts and deleted the guest account and now I have only one admin account with a password. The XP laptop now connects if I enter that username and password. Make sure also that your folders are shared correctly in win7; sometimes sharing with the homegroup is not enough and you have to add 'everyone'. Oh and take into account the security implications of sharing if you're permanently connected to the internet. Sandman30s (talk) 23:10, 25 August 2010 (UTC)
 * Are you sure you have an account on the Windows 7 computer which is doing the sharing with a password and access (permission) to whatever is being shared? Also have you made sure the sharing is allowed thorough the firewall of the Windows 7 computer? You can also try 'manage network passwords' on the Vista laptop (that is trying to access the Windows 7 computer and failing) and delete all credentials for the Windows 7 computer. 10:38, 26 August 2010 (UTC)


 * That's incredible - switching the Guest Account 'off' on the Win7 laptop fixed it. I'm staggered. Thanks a lot! -- KägeTorä - (影虎) (TALK) 13:50, 26 August 2010 (UTC)

I hate a dialog box.
You know what modal dialog box is the worst? Yes, you do. It's this one, from Office 2007, which is triggered when you close a document under certain circumstances:

There is a large amount of information on the Clipboard. Do you want to be able to paste the information into another program later? To save it on the Clipboard so that you can paste it later, click Yes. To delete it from the Clipboard and free memory, click No.

Since this is 2010 and I am no longer using a floppy disk for my primary storage medium, I would like to never see this dialog box again. Does anyone know a way to configure Office to banish this thing? Comet Tuttle (talk) 17:23, 25 August 2010 (UTC)


 * Microsoft Office uses a different clipboard than the default Windows clipboard. (Actually, they're "connected," see the Clipboard technical documentation).  Office Clipboard offers a set of enhancements to the regular Windows clipboard (copy/paste) architecture - things like the ability to store a history for the copy-paste buffer, or the ability to paste the "same content" as multiple formats like plain-text, formatted-text, HTML, or as a raster-image, depending on where you paste it.  (Well, some of that is actually handled by the Windows clipboard - but Office "enhances" these features).  Office Clipboard shuts off when Office shuts off - so it's asking you politely whether you want to "convert" everything back to the un-enhanced version.  Your options are:
 * Never use office enhanced clipboard: disable it permanently by modifying your registry.
 * Turn off the enhanced clipboard for one session
 * Always use the Office Clipboard: run an Office program or component in the background, at all times, so that the enhanced Clipboard engine never turns off
 * Or, live with the dialog-box, because you might want those enhanced features; and you have to make a decision about what to do with clipboard buffers when you turn Office off.
 * Nimur (talk) 18:39, 25 August 2010 (UTC)


 * I don't really get the relevance of floppy disks Nil Einne (talk) 23:45, 25 August 2010 (UTC)


 * Nimur &mdash; thanks for the links. The nag dialog actually doesn't seem to only trigger when I exit all Office apps, but also each time I just close an individual Excel window while Excel is still running.  Nil Einne &mdash; I had assumed that this awful dialog box was a holdover from the 1980s when I remember some Macintosh apps that would ask you this question in an effort to save some memory in order to reduce the amount of floppy disk thrashing as resources would swap in from disk and be purged because, perhaps, the Clipboard was taking up 50K ... never mind.  In any case I think the comment about saving memory on the dialog box I'm complaining about now is positively quaint.  Comet Tuttle (talk) 00:09, 26 August 2010 (UTC)

There is any Web crawler program that is very easy to use and free?
There is any Web crawler program that is very easy to use and free? To be more specific I want to download every song page of pandora website (example: http://www.pandora.com/music/song/anberlin/whisper+clamor ). If possible i want want one that is able to download only the pages that fit on the search, like downloading only the pages with "electric rock instrumentation" on it or not downloading pages with "folk influences" on it.201.78.204.144 (talk) 18:03, 25 August 2010 (UTC)
 * HTTrack is free and easy to use. It can't filter based on page content, but you can specify url scan options 1230049-0012394-C (talk) 18:28, 25 August 2010 (UTC)

#1 Google hit
I've been patrolling new pages, and a couple of times I've Googled the titles of a brand new page, only to have the new Wikipedia entry returned as the first hit. How is that possible? Exploding Boy (talk) 23:57, 25 August 2010 (UTC)


 * Google picks up on new articles very quickly - much more so than could be done with a web spider. It's very likely they follow either (or both) the recent-changes or new-articles RSS/ATOM feeds. -- Finlay McWalter ☻ Talk 23:59, 25 August 2010 (UTC)


 * For example, they picked up on Marti Melville within 2 minutes of its creation. -- Finlay McWalter ☻ Talk 00:03, 26 August 2010 (UTC)


 * OK, I guess I don't understand Google. I thought position on the results page was based on payment (sponsored links) or number of links from other pages.  Exploding Boy (talk) 00:06, 26 August 2010 (UTC)


 * Yes, that's roughly how PageRank works. But the specifics of how Google implements it, and other factors that they put into the equation, are trade secrets.  They change this stuff, and don't publicise how it works, as doing so would be a great boon to spammers and search-engine manipulators.  It seems that they give Wikipedia articles a boost just because they're on Wikipedia; Marti Melville has no intrawiki links here, and (having never been created before) surely has no inbound links. I guess the effect of the secret-wiki-boost would be most noticeable when, as with the MM article, the rest of the pages that Google knows about the subject don't appear to be very high PageRanking themselves. -- Finlay McWalter ☻ Talk 00:14, 26 August 2010 (UTC)


 * Also note that, more than 10 minutes since you deleted MM, it's still in the Google results (in fact it's gone up, from about #10 to #5, perhaps due to you and I doing the search for him a few times). So they seem to follow new creations more aggressively than deletions. Perhaps (hint hint, hypothetical Googlers!) Google should debounce things better, wait for a 2nd editor to edit a new article before it shows up in the search, and follow the deletion stream more vigorously. -- Finlay McWalter ☻ Talk 00:25, 26 August 2010 (UTC)


 * My impression is that PageRank is based not just on the incoming links to the specific page but also in part to the quality of the links to the entire domain. That latter factor will be available immediately even for new pages.  Hence there is a sort of default page rank assigned to new pages based on the quality of the domain hosting them.  Obviously Wikipedia does very well in that regard.  Of course, it is also possible that Google has special rules for dealing with Wikipedia in particular.  Dragons flight (talk) 09:16, 26 August 2010 (UTC)


 * Incidentally they also seem to be pretty smart about parsing our articles. Note how the summary they show for articles starts with the real text - they strip out templates, see-alsos, infoboxes, etc. I'd be surprised if they didn't have one or two engineers devoted full-time to making sure their retrieval and processing of Wikipedia (and probably other MediaWiki-running sites like Wikia) is as good as it can be. -- Finlay McWalter ☻ Talk 00:18, 26 August 2010 (UTC)


 * (Sorry to perseverate on this, but heck it's interesting) they're much better than they used to be about lowering the ranking of mirrors of Wikipedia content. There was a time, maybe three or four years ago, when a certain fact-aggregation website sometimes returned a higher position that Wikipedia, even though it chiefly regurgitated old Wikipedia content. The mirrors are still there, but they're lower now. -- Finlay McWalter ☻ Talk 00:33, 26 August 2010 (UTC)


 * Sure, they're good at what they do. I find it better to use Google to search Wikipedia, rather than Wikipedia itself, e.g. - even for userspace stuff . Then again, they have a budget of $9001M and we have about 10c.  Chzz  ►  01:40, 26 August 2010 (UTC)


 * This is indicative constant improvement in the relevancy and speed of the google index. For more on their latest algorithmic tweaks,  see Google Caffeine, which could explain this. --rocketrye12talk/contribs 21:53, 27 August 2010 (UTC)