Wikipedia:Reference desk/Archives/Computing/2009 April 10

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

April 10

Compressing my old music

I was thinking just now about how valuable my music collection is to me. I've been keeping it exclusively on an external hard disk. and it is taking up about 60 GBs. I've worked very hard to put this collection together and I'd be a tad destroyed if I were to lose it (even just the Table of Contents from the drive). I have another 250 gb drive that I use for media.

I'd like to do a backup of this music on my media drive. What is the best compression format for a collection of (mostly) mp3s 60 GB in size? NByz (talk) 02:30, 10 April 2009 (UTC)

Mp3s are already compressed; further (lossless) compression is unlikely to yield a significant reduction in size. Your best bet is probably just to copy the entire collection to the second disk. – 74  02:39, 10 April 2009 (UTC)

Broken XP

A friends computer encountered a stop error(0x00000021a). Anyways, the only way to get access to his hdd was to hook it up to one of my computers as a slave. I removed any spyware that I could find off of it and replaced winlogon, csrss, msgina. At any rate, none of these things helped. My best guess is that there is something nasty loading up at boot or something wrong with the registry, thus, this is my question: is there any way to access the registry or modify the startup programs for the copy of windows on the drive when it is set up as a secondary drive? (Obviously, attempting to do this the usual way only allows me to change the settings for my primary drive.) Thank you for any help:) Phoenix1177 (talk) 04:15, 10 April 2009 (UTC)

Have you tried safe mode? Or an on-top reinstallation of XP? F (talk) 09:59, 10 April 2009 (UTC)

It won't go into safe mode, that's why I had to hook it up to a second computer. I, unfortunately, don't have any of the windows disks to do a reinstallation. If I could just edit the registry of it from the second computer, I could fix it; I'm almost positive what the problem is. Unfortunately, I don't know how to get at the registry since its not the registry being used by Windows on the computer the drive is a slave to. Is it possible to edit by directly manipulating the files the registry is stored in? 66.202.66.78 (talk) 10:19, 10 April 2009 (UTC)

In regedit, highlight HKEY_USERS or HKEY_LOCAL_MACHINE and go to File -> Load Hive. Then browse to the second hard drive and load up some hives (on XP: Windows\system32\config). Regedit will ask what to mount the hive as, make sure to use a name that won't conflict (like deadsystem_software and deadsystem_system if you load the software and system hives). Washii (talk) 22:45, 14 April 2009 (UTC)

I would try putting the hard disk back in the original computer, and booting it with BartPE or some other "rescue" boot disk, and see if that allowed you to eventually fix the problem. See also List of live CDs. The free virus program Avast! allows boot-time virus scanning, but I do not know if it would work with a "rescue" disk. When I tried different memory in my computer, and indavertantly swapped the order of the memory in the slots, that produced similar problems. Another problem where I could only boot into safe-mode was caused by a defective video driver or videocard - disabling it solved the problem. 78.147.135.185 (talk) 22:52, 11 April 2009 (UTC)

May i suggest teh of teh live cd BackTrack? —Preceding unsigned comment added by 64.30.106.235 (talk) 06:32, 12 April 2009 (UTC)

Disc space

Can I increase the space of an already created disc drive?I did an idiotic thing by creating a small 4.87GB drive for storing the system files and the program files.Now it is almost full leaving a paltry 32MB.I am not able to install MS office since it requires 100 MB space on the drive in which the system files are stored.I think it is compulsory that it should load its files only on the system files drive.Is there any other way out? —Preceding unsigned comment added by 59.165.84.9 (talk) 04:16, 10 April 2009 (UTC)

Your question appears to make sense, but actually doesn't. I seriously doubt you created a disk drive. Did you create a partition on a drive? If so, you can resize partitions (there are many programs that do it - just Google for "resize partition"). If you want to change the size of an actual drive, you are out of luck. That will require taking it apart and replacing all the insides - which is far more expensive and difficult than just purchasing a larger disk. Do you want to install a second drive? That will help. However, Windows will not automatically use it. You need to manually move files to the second disk. However, only move things such as music and movies. If you move executable programs, you will cause problems. The registry knows where the programs currently reside and will not realize they just moved to another drive. If you can make your question clearer, please do. Then, someone can provide a better answer. -- kainaw™ 04:23, 10 April 2009 (UTC)

Apologies to Kainaw, above, but while your question appears not to make sense, actually it does. Or anyway it does if your "32GB" is, as I suppose, a mere typo for "32MB", and if, like many people, you are a bit hazy about the distinctions between partitions and logical drives. Still, it's likely that you created too large an additional partition (whether primary or extended). Yes, you can get software that will adjust the boundaries separating partition from partition, decreasing the size of what's unnecessarily large and adding this to what's too small. It's a very long time since I've done this but if my memory is working right this might involve some time and disk thrashing. Typically there's an awful lot of junk in C: and also a lot of material in C: that you do need but that could go elsewhere. (For example, where are your browsers? Where's your mail program, and where are your mailboxes? On my sole surviving Windows machine, whose C: drive is roughly the same size as yours, they're all in E:, not C:. And, before I deleted it [because I never used it and wanted the space], MS Office too was running off a combination of C: [a bit] and E: [a lot].) Incidentally, OpenOffice requires little or no space on C:. -- Hoary (talk) 05:16, 10 April 2009 (UTC)

Partition editor —Preceding unsigned comment added by 82.43.88.87 (talk) 06:48, 10 April 2009 (UTC)

Yes, and for junk removal see nLite and vLite and XPLite. -- Hoary (talk) 09:11, 10 April 2009 (UTC)

Before you go messing with the partitions, it would be a good idea to back up the entire system including the 4.87GB drive. Pay particular attention to your documents, photos, music, emails, and internet bookmarks - ie. the stuff that is real hard (or impossible) to get back if it gets deleted. Astronaut (talk) 15:01, 10 April 2009 (UTC)

WLM & dll problems

i can't sign into my WLM (windows live messenger) accounts no matter what client I'm using (WLM 2008 8.5.1302.1018 and Pidgin 2.5.5). I get error code 80048820. I tried the steps on this site but when i try to register wintrust.dll i get the error 0x80070005 and am told by my computer to do an internet search to fix it. well um, screw that; an internet search only comes up with forum posts of people who have the same problem but don't know how to fix it. cursory googling tells me this might have something to do with 1. cookies, 2. windows defender, 3. permissions or possibly all of the above. can anyone help me? i am an MSN addict I need my freaking MSN. 99.245.16.164 (talk) 04:28, 10 April 2009 (UTC)

Using sed to remove characters

I would like to have sed remove a specific character. The twist is, I need to be able to remove it only when it appears in a certain context. For example, I want to remove dashes from numbers in a document: [Document 000-00000-000-000000-00], but only when they appear in a certain context (with brackets and the word document before it, for instance). I know how to do this if I know the number of dashes, and I know how to do it if I wanted to remove all of the dashes in the document (tr -d '-'), but I don't know how to make it work if I want to remove all dashes that fall inside of some standard formatting, without actually knowing where and how often the dashes will occur. Any ideas? Other basic unix utilities that will do the same job are fine, but I don't want to get any more complex. Thanks. Shadowjams (talk) 05:37, 10 April 2009 (UTC)

One way would be to write some code to remove the first dash found inside those brackets, then put that code in a loop and run it as many times as the maximum number of dashes which might occur on such a line. This isn't the most efficient approach, but, if performance isn't a problem, it might be the simplest. StuRat (talk) 14:31, 10 April 2009 (UTC)

This'll do it:

perl -pe 's/-//g if /^\[Document.*\]$/'  file.txt

--Sean 15:23, 10 April 2009 (UTC)

Thanks. Perl works perfect. I knew I could put the whole thing inside of a loop, but I wasn't sure how to do that loop in bash and have it exit after it had finished, but only after it had finished removing all of the dashes. Come to think of it, perhaps I could use grep to do that search, but it looks like that perl will be more efficient. Shadowjams (talk) 23:33, 10 April 2009 (UTC)

Here's one in Sed:

sed '/^\[Document [0-9-]*\]$/s/-//g' file.txt

or if you want it to work with [Document] entries embedded in text, not just on a line on their own, you can use a loop though you have to split the sed script across multiple lines:

sed '{:a
s/\[Document \([0-9]*\)-\([0-9-]*\)\]/[Document \1\2]/
ta
}' file.txt

Awm (talk) 22:14, 11 April 2009 (UTC)

Computer Languages

I want to build a base strong in computer languages. I want to learn Python;Lisp;Perl;Javascript;Java. Can someone suggest me books available in India for deep knowledge of these above mentioned languages? —Preceding unsigned comment added by Ingineered (talk • contribs) 06:57, 10 April 2009 (UTC)

I am not a programmer. Nevertheless.... There's learning, and there's deep knowledge: they're different. In your position, I'd start by learning the basics of one language, and for that purpose I'd ask about a good introductory text (which might be one that allows for deep knowledge). With my elementary but functional understanding of one language, I might want to explore the language further or I might want to move on to another language. If you don't know which language to start with, I'd choose the one that's most likely to let you achieve something that you want to achieve in the short term: that will be a powerful incentive for practicing. Of course, there may be texts that are good introductions and also examine theory and go into advanced areas, but they're likely to be big, so they're likely to be expensive, and if after you have consumed 20% of a large, wonderful and expensive book you decide that the language is not the right one for you, you may wish you hadn't spent all the money. -- Hoary (talk) 09:25, 10 April 2009 (UTC)

My advice is to start actually writing code in a single language. It makes learning far more rapid and deep. Write larger and more ambitious programs. The point of a university education in computer science with an emphasis in programming isn't to teach graduates all the details about every computer language, but to teach them how to program. Once they know how to program in one language, it only takes 2 or 3 weeks for the graduate to learn 90% of what he needs to know about any new language. Anyway, learn to code by writing code. This link is Microsoft's free Visual Studio Express versions; one is for BASIC, one is for C#, and one is for C++. Tempshill (talk) 16:45, 10 April 2009 (UTC)

www.infibeam.com/Books/ is an India-specific online bookstore. If you search on the various programming languages you are interested in, these appear to be the exact same books I see in the UK and US - so I don't think there are special recommendations required here. Personally, I like the O'Reilly books - (they all have white covers with line drawings of animals on them) - those are great for delivering the information with the minimum of extraneous junk - and they remain good reference material, long after you've basically grasped the language.

If you are already a programmer (ie if you already understand one language pretty well) - then this is certainly the way to go about picking up these other languages. However, if you are not already pretty competent with at least one language - then I strongly agree with Tempshill that you should concentrate on just one language because first you have to learn how to program...and in a sense, that's independent of the language you choose. So pick one (I'd suggest Java) - study it and write as many programs as you can...the only way to get good at programming is to do a LOT of it!

After you've been writing programs with one language every single day for at least a year - you can call yourself a "programmer" and start looking at learning a second programming language. The second language you learn will be tough because while some things will translate pretty cleanly from one language to another - some things have to be "unlearned" because they are different or do not apply at all for your second language. At this point, you don't know what things are true of pretty much all languages - and which were unique to the first one you just happened to learn. However, after a couple of months - you'll have that second language down pretty good and you can start looking at more and more additional languages.

Eventually, you'll be able to learn a new language in just a couple of days and become fluent in it after just a few weeks of practice - but that's mostly because you'll be saying to yourself: "Oh! That feature is just like JAVA and this other feature is just like Python..." etc - so the more languages you know, the easier it is to learn another. I have literally lost track of the number of languages I know. It has to be more than 30...but then I've probably forgotten a good number too. There comes a point, where you really don't bother to learn new languages unless you absolutely need them. Sure, I know enough 'Python' to get by - but since I can do nearly everything I want in C++, PHP and JavaScript - I very rarely need to serious amounts of work in Python, Java, Perl, etc. Hence, my need to retain details about those languages is much less since I pretty much only need those skills for debugging and extending existing programs written by someone else. This weekend I learned MEL (so I can write scripts in Maya) - I didn't even bother with getting a book - the online documentation was enough. I doubt I'll ever write more than a few thousand lines of MEL...but it didn't take much to pick it up and there are a few tasks I have to undertake for which it's the only language available...so I have to learn it. SteveBaker (talk) 07:21, 12 April 2009 (UTC)

Hoary has it just right if you're serious about programming. And SB also has it right, use mastery of a single language to leverage yourself toward deep knowledge.

I disagree with the recommendations, however. And the reason I do so is because of a superb introductory text. K N King's Modula-2 is superb, nearly encyclopedic (fitting, eh?) and about a language it's possible to learn completely. No one knows everything about C++ (and certainly no one uses all features with facility). Same with Java, and the Microsoft attempt at an equivalent, the .NET projects. Nevertheless, Modula-2 includes a sort of elementary object oriented arrangement (but much easier for a beginner to follow than the full blown business in some languages), and modern program structure provisions as well. Unlike some programming languages, Modula-2 encourages clear thinking about programming, and makes opaque programming style difficult -- you must go out of your way, while in some languages, poor programming style is not only allowed, but is tempting and common. There are free compilers available, perhaps especially for the xxxBSDs or one of the Linux distributions. All have advantage of no cost, and the further advantage of reduced opacity, as in Microsoft's various editions of windows. An important consideration for someone in a learning mode.

So get King's book, start at the beginning and go to the end, answering every exercise, and making a serious attempt to answer even the advanced exercises. It will probably take something like 6 or 8m months, if you stick to it with discipline. King will provide much support whilst learning.

Next, or simultaneously while finishing the last 1/4 or 1/3 of King, read the original edition of Software Tools by Kernighan and Plauger (NOT the Pascal edition), and do all the exercises in it as well, but in Modula 2. This will force you to confront differences between languages in a systematic way, a useful thing for a new programmer to learn. Software Tools is superbly written and is a great learning tool as well. You'll be spoiled for the usual not-so-high quality of most computing books, unfortunately.

King is at Georgia Tech in the US, and both Kernighan and Plauger were at Bell Labs (the original, legendary one) when they wrote their book. In fact, Kernighan is generally credited with the name Unix, and has helped write several other excellent books on programming.

There are likely to be used copies available at one or more of the big Internet booksellers, so cost can be somewhat reduced as well.

You may want to look up E W Dijstra's savage, and hilarious, comments about various programming languages. It was one of EWD series (perhaps nr 485?; Google for BASIC and brain damage and EWD -- it'll turn up), and all are available on the Net. His opinions are blunt in this little note, and not very politically correct. But, though funny, they're more accurate than most will publicly acknowledge. A remarkable man and an important computing pioneer (he invented semaphores, possibly modern style operating systems, and ...). ww (talk) 10:54, 19 April 2009 (UTC)

Files not played (Q moved from entertainment desk)

Hello. When i open a .avi movie with any of my players, i am just getting a black screen. The audio is coming fine, but no video. Please help me. Files that have played well previously are also now facing the same problem. —Preceding unsigned comment added by Rkr1991 (talk • contribs) 07:43, 10 April 2009 (UTC)

Have you tried VLC Player? That seems to play .avis pretty well. --Richardrj ^{talk email} 08:05, 10 April 2009 (UTC)

I have tried all players in the world! Including VLC player. The files played very well previously are also facing this problem. What do i do? —Preceding unsigned comment added by Rkr1991 (talk • contribs) 08:11, 10 April 2009 (UTC)

Well, it sounds to me like you have a codec missing. I don't know much about this, but until someone more knowledgeable comes along you might want to have a look at Video codec. --Richardrj ^{talk email} 08:30, 10 April 2009 (UTC)

I don't think it's likely be a codec problem, especially if they were playing before and now aren't. VLC plays pretty much anything, so if that's not working either then I'd look somewhere else for the problem, perhaps some sort of graphics card error? I don't know. —Preceding unsigned comment added by 82.43.88.87 (talk) 13:31, 10 April 2009 (UTC)

I'd recommend downloading GSpot and opening the AVI file in that. It'll show you what codec the video is using (even if you don't have it installed) and with that we'll be able to work out where to go next. ZX81 ^talk 15:12, 10 April 2009 (UTC)

Facebook ?

If I am logged into Facebook, search for somebody by name, and click on the profiles that come up to get more details, will those people know that I have been checking out their profiles, even for example under the "other people you might know" section ? Don't worry, not up to anything nefarious, just want to avoid embarassment ! Thanks ! --41.15.58.173 (talk) 09:45, 10 April 2009 (UTC)

No. F (talk) 09:57, 10 April 2009 (UTC)

Short answer; No. Long answer; Maybe. To elaborate on the long answer; It is possible for that person to find out who has looked at his profile, But he/she must have access to FaceBook's logs or your personal computer's logs. several ways this person could obtain this, either by directly hacking FaceBook's computers or by hacking your computer. Another way would be if your computer is infected with malware, where this malware uploaded your browsing history to some website, and that website desiced to try and blackmail you with that information... highly unlikley but it is not infeasible. If you are really parenoid then i'd suggest you try and go though a proxy server... or go to a puplic libarey ... or take a laptop up to an open (or, using aircrack, closed) network and look from there... —Preceding unsigned comment added by 64.30.106.235 (talk) 06:26, 12 April 2009 (UTC)

What they mean to say, is no, unless they have the technological knowhow to hack into facebook and find out. —Preceding unsigned comment added by Yakeyglee (talk • contribs) 00:41, 13 April 2009 (UTC)

Is it possible to have a dynamic ip for one wireless network and a static ip for an other without having to change the settings each time you connect.

Well it's basically all in the subject line. I want a static IP in one wireless network and dynamic ip in the other ones, how do I do it guys? Bastard Soap (talk) 13:32, 10 April 2009 (UTC)

You could write shell scripts that configure the network interface. The details depend on your operating system; here are samples for Windows[1]. Write two .bat files, one for static and one for DHCP, and place shortcuts to them on your desktop for quick access. For more instructions and other operating systems google is your friend: [2]. 88.112.62.225 (talk) 08:52, 11 April 2009 (UTC)

Do you use Windows? Look up Alternate Configuration in Help. You change the settings in TCP/IP on your network adapter. Imagine Reason (talk) 16:14, 16 April 2009 (UTC)

Reusing laptop DVD drive

I recently dismantled a broken laptop for spares. One thing I thought would be useful was the CD/DVD drive, which I imagined I could reuse in an external enclosure. Unfortunately, the connector on the back of the drive doesn't seem to be compatible with the external enclosures I've seen in my local computer accessories or on eBay. The connector looks like this, but all I can find is enclosures with standard IDE or SATA connectors. Does the connector on the drive have a specific name that I should be searching for?

Ideally, I'm looking for a slim enclosure that doesn't need an external power source - ie. it draws power via the USB connection. I've seen such a thing for sale, but it already has a drive fitted and is quite expensive. Does anyone have any hints on how to find just the enclosure? Thanks. Astronaut (talk) 13:34, 10 April 2009 (UTC)

Sorry, don't know what connector that is and can't quite make out the model number on the label in order to Google that model number and find out what sort of connector it is. One point that comes to mind, though, is that new DVD-ROM drives cost US$30 currently, and DVD±RW drives cost only US$10 more, and any enclosure you buy is likely to cost around US$25 at least ... so this used drive is probably not worth converting into a working drive. Tempshill (talk) 19:41, 10 April 2009 (UTC)

I'm pretty sure the connector is standard across many (or maybe all) laptops. I just don't know what it is called.

I know I can buy a new external DVD drive in an enclosure for around £40, but this is actually for a friend who is on a very limited budget. I thought that since I can buy an empty external enclosure (IDE or SATA) for about £10, I should be able to get a similar thing to take the old laptop DVD drive for a similar price - and that is much closer to the budget my friend has in mind. Astronaut (talk) 00:49, 11 April 2009 (UTC)

Do these look like it: [3][4]. 88.112.62.225 (talk) 08:46, 11 April 2009 (UTC)

Yes, that's exactly what it is like. So it's a 50-pin JAE.

I'll try to find an external drive enclosure for a slim DVD drive, or failing that get a JAE - IDE adapter and try a regular IDE drive enclosure. Astronaut (talk) 11:33, 11 April 2009 (UTC)

If you do get a JAE to IDE, you still don't have USB. IDE to USB converters run about $20 to 30 US, depending on how fancy they are. You'd need a JAE to IDE, then an IDE to USB, and these do NOT draw the power through the USB. Then you need an enclosure of some kind. It is going to come to a lot more than getting a new one. Really the cheapest way to put this drive back into service is to mount it as an internal drive, with only a JAE to IDE converter cable needed, and maybe a splitter for one of the existing Molex power cables. When you mount it, cut the drive bay cover plate to fill the gap left by the DVD being a slim one.KoolerStill (talk) 05:11, 19 April 2009 (UTC)

Linux devices

I'm quite a linux newbie and one thing has been puzzling me about the devices in /dev. If I connect a device it needs to be mounted using a command like mount /dev/??? /mnt/somewhere When I connected an external IDE disk drive via a USB cable, I looked up and down the many devices in /dev and nothing hinted at which device I should mount. It was only when I looked in /etc/fstab that a comment line suggested to me that external USB drives were /dev/sda1, /dev/sda2, etc, and I successfully mounted the drive. But, is there a way I can tell which device in /dev is the one to which my physical device is actually connected to? Astronaut (talk) 15:21, 10 April 2009 (UTC)

There is some variability, but there are guidelines about how things are listed in /dev . IDE hard drives and CD drive will generally be listed as hda, hdb, etc and the number after it is the partition number. SCSI drives are listed like sda, sdb, etc (USB disks will also be listed like this because Linux tends to use SCSI emulation). Other devices have handy shortcuts like /dev/audio, /dev/cdrom, etc. Have a look at Device_file#Naming_conventions for the specifics. Freedomlinux (talk) 15:41, 10 April 2009 (UTC)

SATA drives are also listed as sda, sdb, etc.  Buffered Input Output 12:52, 13 April 2009 (UTC)

The easiest way is to use dBus to detect when external drives are added/removed. The hardware abstraction layer (HAL) sends a signal along the system dBus which all programs can see (this is how Nautilus realises usb sticks/cds have been inserted). --h2g2bob (talk) 17:33, 10 April 2009 (UTC)

dev has loads of devices other than just disk drives. The naming convention in MOST linux setups is to use sd for sata, then a sequential letter for a drive, and a sequential number for the partition. So sde1 is the fifth sata drive, partition one. IDE drives are usually hd with the same drive and partition sequencing. The simplest way to determine what node a device is assigned to is to take a look at "dmesg" which is a copy of the recent kernel messages. Just type dmesg at the shell, and it might help give you some hints as to the device. Shadowjams (talk) 02:38, 11 April 2009 (UTC)

See also udev, which (when you have it) reduces the confusion by dynamically creating the device files that correspond to what currently exists. It's still a bit inobvious how to determine which device is which, but it helps; the /proc and /sys information directories can also be of use, although there's a lot to sort through in them. --Tardis (talk)

When I'm curious about which device file corresponds to something I just plugged in, I run "tail /var/log/syslog" (on some systems syslog.0). Usually some diagnostic messages about the device being detected will be printed there, and they usually mention the device filename. Indeterminate (talk) 10:43, 12 April 2009 (UTC)

Failed emails which I never sent

Logging into Windows Live Messenger today, I find myself with 19 emails. Surprised, I check them to find that they are delivery failure notices. Every invalid address on my contact list is contained within these 19 emails which tell me the message I sent failed to reach them. My first thought was that someone obtained the email addresses (presumably by somehow obtaining my password) and emailed everyone. This would seem to be confirmed by the list of addresses at the bottom of each failure notice; the list is of every address the email was sent to, and between them, the 19 emails contain every contact I have.

However...

I've contacted several people who are on the lists at the bottom of the emails. None of them receieved the email.

Can anyone explain this? Vimescarrot (talk) 18:19, 10 April 2009 (UTC)

I get these phantom failure notices all the time, and have for years. There are a couple of possibilities. (a) Your system has been infected with a computer virus or some other malware that read your address book, and then sent copies of itself to everyone on that address book (or maybe just sent them Viagra ads). (b) A friend of yours has been infected with similar malware, and the malware signed each outgoing e-mail with your e-mail address, as though you personally had sent it. (In case (b) here, your computer itself was never involved; the malware just found your e-mail address in your friend's address book and thought it would be fun to pretend you had sent the e-mails.) I would scan your computer with two anti-malware scanners, for safety, and if nothing is found, then just ignore these phantom failure messages. Tempshill (talk) 19:32, 10 April 2009 (UTC)

(edit conflict) - Spammers will often "spoof" a valid email address in hopes of getting more views once their own accounts have been flagged as a blacklisted account. That basically means - someone else used your email address as the return address for some spam they sent out. You can try to track IP numbers and report them to various ISP's, but in the long run it's going to be a lot of work with little result. I just delete those messages myself. — Ched :  ?  19:35, 10 April 2009 (UTC)

Don't forget that these email messages contained my entire contact list; somehow, they had access to my account; they weren't just using it as a return address. I scanned four days ago and did nothing out of the ordinary since then; is it really worth scanning again? Vimescarrot (talk) 19:43, 10 April 2009 (UTC)

If you just saw the failure messages today then maybe it's scenario (a) above. Yes, scan again; it doesn't hurt. And scan with an alternative program, too; all malware scanners have different approaches and find different things. Tempshill (talk) 21:55, 10 April 2009 (UTC)

Executables files on Linux (ELF)

Resolved

I was just wondering if ALL executable files that run on Linux (Ubuntu, Fedora, Debian, etc...) are ELF files? PS: I am not counting with scripts (nor stuff like wine to run .exe files) Thanks -- SF007 (talk) 20:32, 10 April 2009 (UTC)

No, they could also be a.out or a number of others. If you look at the files called binfmt_XXX.c here, each one implements a different binary format. --Sean 22:27, 10 April 2009 (UTC)

In practice, however, a modern system will only have ELF binaries - the a.out article says that Linux's a.out implementation was unsuited to shared libraries, and was phased out fairly early on in Linux history.

The binfmt-related kernel config options show what's available, and include the message:

"Linux used the a.out formats QMAGIC and ZMAGIC until they were replaced with the ELF format. The conversion to ELF started in 1995. This option is primarily provided for historical interest and for the benefit of those who need to run binaries from that era."

An interesting point is that "binfmt_script" is what implements shebang handling, and binfmt_misc allows binding various kinds of magic to interpreters or virtual machines, so the kernel can execute a Perl script or Java class in much the same way as an ELF or a.out binary, blurring the distinction made in the original question. - IMSoP (talk) 23:04, 10 April 2009 (UTC)

Cool, thanks to all. SF007 (talk) 03:13, 17 April 2009 (UTC)