Computing desk
< March 1	<< Feb \| March \| Apr >>	March 3 >

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.

March 2

Key exchange with no shared secret?

Diffie-Hellman key exchange requires that the parties first be able to establish a shared secret over a secure channel. Is there any way to perform key exchange if the parties have never had and will never have a secure communication channel? Neon Merlin 00:15, 2 March 2010 (UTC)[reply]

Actually, Diffie-Hellman allows a shared secret to be established over an insecure channel. decltype (talk) 00:17, 2 March 2010 (UTC)[reply]

No, you're wrong, Diffie-Hellman does not require that the parties establish a shared secret over a secure channel first. THat would be called Symmetric-key cryptography, and it is what people did for thousands of years before Diffie-Hellman and RSA was discovered. The the whole point of the Diffie-Hellman key exchange is that even if you hear every single thing communicated, you will not be able to figure out the key if you're not one of the parties.

There is a big problem with Diffie-Hellman though: the man-in-the-middle attack. If we ask "assume you can communicate over a insecure channel where the eavesdropper can listen in but is unable to change the messages", then Diffie-Hellman, or even RSA (you could just send the other party your public key) pretty much completely solves that problem.

However, that's not the real world. People often can change the message in transit, so then you need to authenticate your messages. This problem has absolutely nothing to do with Diffie-Hellman, they apply to any kind of public key cryptography, and the solutions are the same. Most commonly, this is solved either by a PGP-style web of trust, or (as is the case with SSL, for instance) by using Certificate authorities. Neither of those methods are perfect, but they're pretty much the best we have. Belisarius (talk) 01:27, 2 March 2010 (UTC)[reply]

Importing TiVo files into Windows Movie Maker

I have a HP Pavilion DV6646us, running Vista 64, 2 gb memory, 160 gb HD. Started using Tivo Desktop to import movies into Windows Movie Maker, edit out commercials, then export to WMV to burn onto DVDs. I've used this Tivo Desktop/Windows Movie Maker combination for a few years now, and have had no problems with it up until a few months ago.

However, now, when I import into WMM, I get a 1:1 aspect ratio rather than the normal 4:3 ratio. The strange thing is, when I simply play the files using Windows Media Player, they look just fine. It seems to be ONLY in Windows Movie Maker that the aspect ratio gets "squared up."

I know it's probably something to do with the codecs, but I'm by no means an expert. I've done Google searches on every combination of "Tivo, Desktop, Windows, Movie, Maker, Aspect, and Ratio" that I can think of, and have come up with very few relevant results, none of which have solved the problem. I found and installed tbe "Cole2K" codec package that was suggested on one site, and I downloaded a "Tivo Desktop Codec Manager" ("TDCM") that I found on the Tivo Community forum. No dice on either.

I also uninstalled TiVo desktop and reassociated the ".TIVO" extension with Windows Media Player, but the video is all scrambled and the files don't play. I then reinstalled an older version of TiVo Desktop, also to no avail.

My main goal here is to be able to import TiVo files into Windows Movie Maker and preserve the existing aspect ratio. That's it. Help! Thanks much, Kingsfold (talk) 03:59, 2 March 2010 (UTC)[reply]

Deleted E-mails

Hi

1.Is there a way to recover deleted e-mails after you've emptied your DELETED ITEMS folder?

2.Is there a way to see whether someone has opened up one of your e-mails without you suspecting? Like opening it and then using the MARK AS UNREAD feature to avoid suspicion.

3.How do you send one of those e-mails where you open it and after closing still looks like it hasn't been opened? Like the one that you get from time to time that says YOUR MAILBOX IS OVER IT'S SIZE LIMIT.

Thanks, NirocFX

41.193.16.234 (talk) 05:31, 2 March 2010 (UTC)[reply]

1. If an only if the email was stored on your harddrive and you know a hell of a lot about file systems, you can get back deleted emails. It is not something can do if you don't really know what you are doing.

2. No. The email doesn't contain any "I was opened on" date. It is just text.

3. The server is refusing to accept the "been opened" change to the email. It isn't the email itself.

I hope that helps. Keep in mind that email is just text. It is a very insecure format designed in the old days when messages didn't need to be highly secure. So far, nothing has been able to replace it and fix the blatant problems with the entire email system. -- kainaw ™ 05:43, 2 March 2010 (UTC)[reply]

(1) If you're on a corporate network, e.g. using Microsoft Exchange Server, then your system administrators may be able to recover deleted messages. AndrewWTaylor (talk) 10:30, 2 March

2010 (UTC)

Thanks, Kainaw and AndrewWTaylor

So you guys are telling me that this is the one department (for as long as it has existed) that hasn't really made progress?

Thanks, NirocFX

41.193.16.234 (talk) 11:07, 2 March 2010 (UTC)[reply]

I'm not sure I'd call what you're describing "progress". Your best way of ensuring no one has read your email and then marked it as unread is to not tell them your email password. As for recovering deleted messages, it sounds like you want a "recycle bin for your recycle bin". At some point you just want the stuff deleted, and indeed internet mail companies hanging on to deleted messages has been an issue raised in the past. TastyCakes (talk) 17:41, 2 March 2010 (UTC)[reply]

Email, as it currently exists, is an old system that hasn't improved much. One of the huge problems with getting back deleted mail is the way in which most programs store emails. They are not stored as single files - so you can search for the deleted file and get it back. They are stored in one huge text file. When you delete an email (and really delete it), it is removed from the text file. Chances are, that process will completely obliterate the email from the hard drive by overwriting it with other emails. So, there is nothing left on the harddrive to get back. As for security, there is none in email. You have to secure the access to email. Use a password. Lock your computer. Don't tell others what your password is. Don't use a stupid password. I showed this at a security talk at a college I used to work for (and this is one of the main reasons that I says "used to work for"). I loaded up the servers passwd file in John (a free password cracker) and ran it while I explained that the school needed to use shadow passwords and they needed to install a password cracker program to force people to use better passwords. While I talked for about 15 minutes, the screen scrolled with hundreds of users using their last name as a password, using "cougar" (the school's mascot), and using the common ones: "god", "password", "fuckyou", etc... -- kainaw ™ 17:58, 2 March 2010 (UTC)[reply]

Thanks TastyCakes, Kainaw

Would you say that it (e-mail as a whole) is flawed (especially with regards to recovery and security/privacy) in a way and that it hasn't been thoroughly taken through it's paces (by the people who created it) to see where it can be improved upon?

Thanks, NirocFX

41.193.16.234 (talk) 08:25, 3 March 2010 (UTC)[reply]

Remove Security Tool

Okay, I guess I clicked on something that I shouldn't have, and so I let this un-godly fiend onto my machine. It comes up with its stupid messages every 15 seconds or so, but far worse is that it won't let me run many programs (including my word processor. Grrr). What's the best way to get rid of it? I'm running Windows 7. I have legitimate antivirus software (McAfee, provided by my University); is that likely to get rid of it? Or do I really have to boot up in safe mode and go after this myself? I'm totally incompetent on these new-fangled computer thingys, so if I do need to go on a search and destroy mission, I'm going to need step by step instructions. I promise to be more careful in the future. Buddy431 (talk) 06:26, 2 March 2010 (UTC)[reply]

Try Security_Tool#Removal. – Elliott(Talk|Cont) 14:33, 2 March 2010 (UTC)[reply]

If that doesn't work, also try combofix —Preceding unsigned comment added by 82.44.55.204 (talk) 15:12, 2 March 2010 (UTC)[reply]

I got rid of it, thanks. That was a lot less ~~painless~~ painful than I was expecting. Buddy431 (talk) 02:53, 3 March 2010 (UTC)[reply]

Saving MS WindowsXP or MSOffice to CD for possible reuse before I delete them

The motherboard on an old computer I had died, so I have an HD with a valid copy of windowsXP on it. I'd like to delete the operating system and have that disk as the secondary HD in another computer. Before I do this, is there any way of saving that operating system to a CD in case I want to restore it to the HD again in the future? (If for example I replace the dud motherboard.)

I also have MSOffice on the HD of the computer I am using now, but intend to delete it. Is there any way of saving that to a CD so that it could be restored again if I change my mind? Thanks Edit:If a CD is not big enough then I could instal a DVD r/w drive from another computer. 78.149.112.209 (talk) 14:52, 2 March 2010 (UTC)[reply]

Installers are (very deliberately) a one-way process (you can't turn an installed program back into install media). It's possible to clone one hard disk onto another, but that uses another hard disk, rather than freeing one, which is what you're trying. This is true for both XP and Office. -- Finlay McWalter • Talk 14:57, 2 March 2010 (UTC)[reply]

That's right — see List of disk cloning software, also. Comet Tuttle (talk) 15:27, 2 March 2010 (UTC)[reply]

However, might it not be that an image of the disk could be burned to DVD? If you are able to resize the partition to less than the DVD maximum limit (by deleting all programs you have install media for and then using something like gparted to resize the partition) and then use some software (I don't know the specifics of that, perhaps gparted can do that too?) to burn an "image file" of your system partition to DVD. Later you can run the process in reverse to burn the image back to the HD if you so desire. Of course, whether it'll boot with a different motherboard, you can't be sure of. Jørgen (talk) 21:27, 2 March 2010 (UTC)[reply]

I don't understand partitions - as far as I know I've never had any on my computers. I imagine making a reusable copy of an installed program would consist of save the program, saving the relevant registry details, and then being able to reinstall them when needed. 84.13.29.241 (talk) 21:35, 2 March 2010 (UTC)[reply]

Partitions are rather easy but that's another question altogether. Basically I know of no software that can make a reusable copy of an installed program in the way described. It would be non-trivial as you would have to know what parts of the Registry to clone, which parts would need to be changed based on settings differences, and other myriad shared files, etc., and things that might vary between different versions of an OS, and so forth. For a trivially small program this would be doable but for XP or Office, I don't think so. --Mr.98 (talk) 01:13, 3 March 2010 (UTC)[reply]

Actually my old HD has both WindowsXP on it and MSOffice, I now realise. So I suppose I could delete everything on it except those things and then clone it, and save the image file to CD or DVD? 92.29.76.9 (talk) 17:39, 3 March 2010 (UTC)[reply]

Can a website disconnect my wireless connection?

I'm sure the answer to this is no. But when attempting to use http://www.old-maps.co.uk after about 5-10 page changes (navigation) - my WiFi connection plunks out. I've been using the internet all day and the effect only happens on this site. It's completely consistent. (Windows XP) (eg try searching for a map, moving around a bit, selecting a different map scale or type..)
How can this be explained? 87.102.67.84 (talk) 17:03, 2 March 2010 (UTC)[reply]

In theory, if your router or wireless card is poorly designed or the firmware/drivers are bad, it would be possibly for it to die as a result of receiving a packet of incorrect size or format. Does the site in question use pure HTML, or does it include Flash programs? TCP should be roughly the same on every site (I wouldn't expect a problem to occur with only one site since the same protocol is used in every case), but a UDP driven site that handcrafts packets might have issues. —ShadowRanger ^(talk|stalk) 17:07, 2 March 2010 (UTC)[reply]

For the record, when I say "incorrect" I don't mean malicious or problematic under normal circumstances. It might slightly violate the best practices for UDP packet construction in such a way that it works perfectly well for wired connections, and it might only trigger packet drops (not wireless connection death) in a properly configured router/wireless card combo. Wireless cards and their drivers expose lots of problems in protocols that were designed and tested entirely within an office LAN, running strictly on wires. And those same bad protocols expose poor error handling in the wireless card and its drivers in turn. —ShadowRanger ^(talk|stalk) 17:21, 2 March 2010 (UTC)[reply]

Just HTML, CSS and Javascript as far as I can tell. Possibly it's using ADF.NET web map control to update the graphics (of maps). Beyond that I haven't a clue.87.102.67.84 (talk) 17:39, 2 March 2010 (UTC)[reply]

Update - like all good gremlins - it appears to have completely stopped doing it now I've asked someone about it. Leaving me even more confused (maybe I'm just going senile).87.102.67.84 (talk) 18:37, 2 March 2010 (UTC)[reply]

It might have just been a coincidence. Humans have a knack for recognizing patterns on limited evidence, even where they don't actually exist. Either that, or the computer is mocking you and will eventually drive you to murder. Take your pick. :-) —ShadowRanger ^(talk|stalk) 18:54, 2 March 2010 (UTC)[reply]

GNU make issue

I have a makefile that WORKS under UNIX using whatever the default make utility is, but not under Linux. Some member functions of one class are said to be "not declared in this scope" when compiling under Linux. In my (limited) understanding of compilation, they are indeed declared, in a header file near the top. I'll add that the file extension that I've used is .hxx, but changing it to something else doesn't seem to help. What might be the problem? Thanks.--Leon (talk) 18:25, 2 March 2010 (UTC)[reply]

It sounds as if you may be missing a library, or a name of the path to the headers is not included int he make file. Graeme Bartlett (talk) 20:56, 2 March 2010 (UTC)[reply]

Your problem is very unlikely to have anything to do with the make utility. "not declared in this scope" is a compiler error message, not a GNU make message. Check out Wikipedia:Reference desk/How to ask a software question for how to ask in a way that permits a more detailed solution. --Sean 17:02, 3 March 2010 (UTC)[reply]

I managed to SOLVE the problem, but only by using another compiler, in this case from intel. I've no idea why it didn't work from g++ (or gcc).--Leon (talk) 21:51, 3 March 2010 (UTC)[reply]

CustomizeGoogle is broken, how do I fix it?

from WP:VPT

I need a new version of http://www.customizegoogle.com/ because Google's new layout doesn't have a Wikipedia link with that anymore.

Would someone who understands how to "convert the CustomizeGoogle extension to a zip file and then edit the files in that" please help me out? My friend says that's the way to get my beloved one-click Google-to-Wikipedia link back.

Anyone who knows how to do this, if you turn out to be local to me (and it's quite possible) then I will babysit your pets, kids, house, or fish as you see fit for a week. Thanks in advance!

Logoth (talk) 06:35, 1 March 2010 (UTC)[reply]

http://www.optimizegoogle.com/ took over the codebase. I've used it for a few weeks (it's identical to customizegoogle) and recommend it to all. :) -- Quiddity (talk) 07:28, 1 March 2010 (UTC)[reply]

Does that work with the new Google layout? They are asking for developers, but they haven't even approved my blog comment after a few days. I'm not sure whether they're working on the new Google layout or not. There is also a SourceForge project and a Google Group.

I made this ranked-choice voting poll so you can pick your favorite search engines to appear on the next version of OptimizeGoogle -- PLEASE VOTE WIKIPEDIA! :D

How do you look at the source code? Logoth (talk) 03:49, 2 March 2010 (UTC)[reply]

XPI and JAR files are just zip files. [Rename the .xpi to .zip and the .jar inside that to .zip,] unarchive and read/edit. See [1] or [2] for more details. (Also, this isn't a good place to discuss non-Wikipedia centered problems. Ask at Wikipedia:Reference desk/Computing if you have further questions :) -- Quiddity (talk) 19:47, 2 March 2010 (UTC)[reply]

Okay, I'm asking here: Can this extension be made to work with the new Google search layout? Logoth (talk) 20:27, 2 March 2010 (UTC)[reply]

IP address

How do I get the IP address of an edit that was made by a registered user? Aaadddaaammm (talk) 20:41, 2 March 2010 (UTC)[reply]

Slave away at Wikipedia with incredibly high quality edits; mediate disputes; get approved as an Administrator; continue to slave away at the thankless tasks of Administrators until you are trusted to be one of the 39 people who have this ability. Or I guess you could ask one of the 39 users. See Wikipedia:CheckUser. Comet Tuttle (talk) 20:43, 2 March 2010 (UTC)[reply]

Or you could ask them on their talk page or via email! Graeme Bartlett (talk) 20:54, 2 March 2010 (UTC)[reply]

To explain why: The point of creating an account is to group your edits together regardless of location and, oddly, to provide an additional layer of anonymity should you so choose. An IP can be traced to a geographical location and sometimes to a specific computer, while your username is not traceable in that manner. The exception to this rule is people with CheckUser (and even they are required to have something resembling the concept of probable cause to perform an IP check). They are trusted with that power so as to rule out abuses of the system, such as one user creating multiple accounts so as to pretend that their edits or point of view has broader support than it does. Aside from scenarios of that type, registered users are supposed to have greater default anonymity than IP users. —ShadowRanger ^(talk|stalk) 21:14, 2 March 2010 (UTC)[reply]

It's not very accurate, but this might be of some use to you —Preceding unsigned comment added by 82.44.55.204 (talk) 21:44, 2 March 2010 (UTC)[reply]

Based on my (signed in) contributions it appears to work by finding when a user either

Signs in and re-signs a previous 'anonymous' post (in Talk)
(or) Accidentally (or on purpose) alters another anon User's comments (in Talk) when signed in.

(It's certainly not going to be effective against someone evading a block of a known username - unless they are very stupid)

If you just wanted the IP to find out what ISP they are using, or which state they are in it would be a lot easier to ask.87.102.67.84 (talk) 23:28, 2 March 2010 (UTC)[reply]

Finding out about an IP is fairly easy here Ronhjones ^(Talk) 23:34, 2 March 2010 (UTC)[reply]

Many IP addresses do not have a very informative whois entry. Sometimes other methods, such as traceroute will provide you with a more detailed location to the user. -- kainaw ™ 23:37, 2 March 2010 (UTC)[reply]

Thanks for the replies - I had no idea the IPs are hidden for registered users. I was just curious about a user, nothing to warrant asking one of the chosen 39. Aaadddaaammm (talk) 09:52, 4 March 2010 (UTC)[reply]

Pre-translated phrase libraries for command-line apps

In GNU/Linux, the strerror() function in <string.h> always returns a description of an error code in the environment language. Do any other GNU/POSIX/C99/etc. standard library functions provide pre-translated text? Neon Merlin 22:34, 2 March 2010 (UTC)[reply]

The key word here is locale (or i18n). strerror returns a locale-dependent result, meaning it tries to present a string in the user's preferred language. strsignal does the same thing for signal names. hstrerror, gai_strerror, regerror, and dlerror are very strerror-like and also can return locale-dependent translated strings. Overall, the C99 and POSIX library functions don't have a lot of strings embedded in them other than error strings. The most interesting one might be strftime, which can do locale-dependent date formatting. 98.226.122.10 (talk) 10:10, 3 March 2010 (UTC)[reply]

Ripping Music from a Gameboy ROM

I've scoured the web for this. I have a ROM of Armorines: Project S.W.A.R.M (legally obtained) for the Gameboy Color. I'd like to get the music off that ROM and into a GBS file. However, I can find ZERO ways to do this. I know people have done it because sites like Zophar.net have tons of GBS music. Can anyone direct me to a guide on how to do this? Or, better yet, a link to an already ripped GBS file for the game? Buffered Input Output 23:43, 2 March 2010 (UTC)[reply]

We're talking Game Boy Sound System. Trouble is, ripping music from ROMs is almost certainly a copyright violation, and wikipedia is not in the business of offering advice on how to do such things. --Tagishsimon (talk) 00:32, 3 March 2010 (UTC)[reply]

I'm not using it for commercial use, just personal enjoyment. And I know it might be a copyright violation, but the game is so old and the company defunct that it would take one desperate lawyer to take up this case. I'm willing to accept that risk if someone could actually answer my question. Buffered Input Output 00:41, 3 March 2010 (UTC)[reply]

Except even if you're willing to take the risk, Wikipedia isn't. By willingly aiding someone in an illegal situation (regardless of how unlikely it is to come up), it's just asking for future trouble. Sorry. ZX81 ^talk 00:49, 3 March 2010 (UTC)[reply]

Ripping music for one's own personal use is not necessarily a copyright violation. (Let's just not ask where he got the ROM from—there are countries in which ROM backups are not illegal.) Anyway, the legal status of such a thing seems hardly clear-cut enough to warrant us refusing to give advice (and ethically, there is really no problem here, which is where I personally would draw the line). (And let's not worry about a slippery slope in such a specific case, eh?)

Anyway, follow the links for the programs at the Wikipedia page that Tagishisimon gave, they should probably help you out. Feel free to consult a copyright lawyer if you are afraid of violating the law. --Mr.98 (talk) 01:06, 3 March 2010 (UTC)[reply]