Wikipedia:Reference desk/Archives/Computing/2008 March 9

Computing desk
< March 8 << Feb | March | Apr >> March 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.


March 9 edit

Windows keyboard transpose edit

Is there a very simple way in Windows (98, XP, Vista...) to transpose the keys on the keypad so that the top row is 123 and the bottom row is 789? -- kainaw 05:10, 9 March 2008 (UTC)[reply]

I use Sharpkeys for all my keyboard remapping needs. It's an easy to use registry tweaker. Works great! — Kieff | Talk 06:14, 9 March 2008 (UTC)[reply]

Dizzler copyright vio? edit

Why does the website Dizzler.com still exist? I’m no expert, but it looks like it’s violating international copyright law. Why haven’t the record companies blown it out of the water? --S.dedalus (talk) 06:43, 9 March 2008 (UTC)[reply]

I poked around it using DownloadHelper, and it looks like the actual MP3 files are on a large assortment of hosts outside the U.S. -- Consumed Crustacean (talk) 08:55, 9 March 2008 (UTC)[reply]
Which doesn't make it legal under the DMCA. See MGM Studios, Inc. v. Grokster, Ltd..--98.217.18.109 (talk) 14:21, 9 March 2008 (UTC)[reply]
The DMCA is not international.. The Pirate Bay is still alive. :D\=< (talk) 19:54, 9 March 2008 (UTC)[reply]
Well, the Pirate Bay has been in the middle of local legal issues for a few years now. Don't rejoice yet. Everybody claims they'll never get taken down, until they do. --98.217.18.109 (talk) 22:25, 9 March 2008 (UTC)[reply]
Anyway, nobody's even trying to say that they're being prosecuted under the DMCA- that's clearly US law :D\=< (talk) 18:55, 10 March 2008 (UTC)[reply]
Lol international copyright law? As much as America would like to think otherwise, the entire world hasn't signed the Berne Convention, and most of the world isn't going to spend one penny on enforcing America's IP for them, even if they have signed it. :D\=< (talk) 19:51, 9 March 2008 (UTC)[reply]
Wow, Froth, who knew you were an expert on international IP negotiations? "Lol" indeed! Actually the countries who aren't parties to international copyright treaties can be counted on a few hands and feet—it has nothing to do with loving the US's approach to things, it has to do with the US having made it a requirement to join the treaties in order to get the big-big money in US trade. The governments of most countries, quite rightly, see their own trade with the US as being more important than whether or not people can most pirated MP3s and movies around without a penny going to the content creator. (I think US copyright laws are draconian, but not for that reason. They're draconian because the copyrights do not expire in a timely manner, if at all. But the idea of copyright itself isn't so bad. People should be able to get paid for what they create.) In any case, Dizzler is registered in the USA, so yeah, US law probably applies. Dizzler itself says it adheres to the DMCA[1], and gives a very dubious legal justification for its action (it uses cases regarding image search engines to justify its own approach, and apparently disregards the Grokster ruling I mentioned above). --98.217.18.109 (talk) 22:25, 9 March 2008 (UTC)[reply]
Oh yeah, that dizzler nonsense is going down hard. Funny they don't hide behind tv-links's defense, which was "oshi don't imprison me please" :D\=< (talk) 18:57, 10 March 2008 (UTC)[reply]

Converting an mp3 into something readable edit

I have a short mp3 that I want to be able to play on an instrument myself. I've gotten most of it by ear, but a few chords are still off. Is there some way I could convert the mp3 into some format that shows roughly what notes are being played? HYENASTE 07:11, 9 March 2008 (UTC)[reply]


Not really, its a very tricky subject. You can play a midi keyboard into cubase or sibelius, and it'll notate it for you. However, it is often a confusing mathematical way of writing the notes, with excessive rests and dotted notes, something which still needs 'translating' into a decent readable score.

As for software that 'reads' an MP3, as far as I'm aware, it cannot be done, and if there is software out there, I doubt how good it is. Radiofred (talk) 14:49, 9 March 2008 (UTC)[reply]

Converting an audio file to a musical transcription is difficult for a computer to do. You can find programs for this purpose, but they really only provide useful results on audio recordings of a single instrument playing no more than one note at a time. For example, you can try the freeware AmazingMIDI, a WAV to MIDI converter, and see how it works for you.
An alternative approach is a program that assists you in analyzing the music, but ultimately leaves it to you to decide which notes best match the audio. I've tried a program called Transcribe! (30 day evaluation, $50 for one copy) that works in this way.
When you try to determine the notes on your own, it can be useful to slow the recording down without changing it's pitch. In Windows Media Player you can go to the View menu and choose Enhancements, Play Speed Settings. Or, if you use Winamp, you can install a plug-in called PaceMaker that can change the speed of files playing. Or you can use an audio editor to change the speed. For example, in Audacity, you can select a portion of the audio, then go to the Effect menu and choose Change Tempo. --Bavi H (talk) 01:03, 10 March 2008 (UTC)[reply]
Generally, most music recordings contain different combinations of sounds and instruments, many of which "overlap" - that is, their sounds share common frequencies. Whilst analysing the frequencies that make up these sounds is a trivial exercise (there are many programs like SPEAR that do it for you), getting a computer to identify the parts you want transcribed is the hard part. There is constant research in this area but as mentioned before, it's a tricky subject. Having said that, recordings of solo instruments are relatively easy to analyse since there is no other interference from other instruments. Chords are hard to determine because typically a musical instrument or sound will be made up of many different frequency components (which define its timbre), so you won't see three unique notes, but a wash of many at different amplitudes.
Unless of course the chord in question is played on an instrument that produces pure sine waves. Good luck! Damien Karras (talk) 13:36, 13 March 2008 (UTC)[reply]

Apache Tomcat? edit

Hi there. In the past I've used the Apache web server (with its PHP module) to run a hobbyist website. I've been without a computer and an internet connection for a while but I shall shortly have both back and will want to set it all up again. However, this time I shall also want to run some JSP / servlet thingies on a separate, private part of the website.

My questions are:

1 - am I right in thinking I can't do the Java stuff with just Apache?

2 - if I replace Apache with Tomcat then will Tomcat be able to do the PHP stuff?

3 - how much of a pain is running both Apache AND Tomcat?

Thanks! --79.64.51.58 (talk) 11:11, 9 March 2008 (UTC)[reply]

I don't know much about this but googling for things like "Apache JSP PHP" turns up a lot of hits. Apparently you can, with a lot of configuration, get Tomcat to do both PHP and JSP (see here); and apparently you can run both Apache and Tomcat at the same time, getting Apache to forward certain types of requests on to Tomcat (see here and here). Hope that helps. --98.217.18.109 (talk) 20:13, 9 March 2008 (UTC)[reply]
Previous poster is correct, you can get Tomcat to do PHP. For a site with more than a few hits the best option is to use a combination of apache tomcat and http server. You can achieve this with mod_jk, there are other ways too but this is the most common way. -- Q Chris (talk) 11:48, 10 March 2008 (UTC)[reply]

decompling proframs wrintten using DBASE and CLIPPER edit

Anyone know how you can decompile programs written some time ago using DBSAE and CLIPPER to get back to source15:15, 9 March 2008 (UTC)

Google "clipper decompiler" and you'll find a lot of programs that do it. --98.217.18.109 (talk) 20:16, 9 March 2008 (UTC)[reply]

Finding out passwords edit

Is it possible for the system administrators/developers of a site to reveal the password of your user account? For example, can the developers of Facebook, Hotmail find out your password given your username? Similarly, can the developers of Wikipedia also find out your password? Acceptable (talk) 16:42, 9 March 2008 (UTC)[reply]

Most sites use a one-way hashing method to encrypt password. That is, they pick the password you choose, hash it to obtain a virtually unique string, and store it in the database. The next time you log, they hash the password you inputted and check if the resulting string is the same as the one stored in the database. This way, they don't have to store your real password (they have disclaimers that, once you forget your password, you lose your account).
So, don't worry, even if their databases are leaked, it is very possible they only store the hashed strings and not your password. -- ReyBrujo (talk) 17:00, 9 March 2008 (UTC)[reply]

Many sites store one-way hashes, and all sites probably should, but there are still plenty who do not. I got into an argument with the developer of a very commonly used forum software about this. He felt that a plaintext password was required so that he could e-mail your password back to you. I think that's chasing a bad reason with a bad reason. It is probably best to assume that your password is recoverable from any site. For what it is worth, the Payment Card Industry standards and OWASP recommendations require secure storage of passwords. There's no way for you, on the outside, to tell if a site is actually following those standards. --Mdwyer (talk) 21:31, 10 March 2008 (UTC)[reply]

But if you had hashed strings you could find out a lot of passwords using rainbow tables. You can pretty easily find out small, simple passwords in such a fashion. When you use very long passwords with non-standard characters, you make it a lot harder. (Unfortunately some sites don't let you use long passwords, even though they are much more secure than the most convoluted short ones against such an attack.) --98.217.18.109 (talk) 17:08, 9 March 2008 (UTC)[reply]
That is true. However, I assumed Acceptable was just wondering whether developers could just say "User X has password Y" without effort. -- ReyBrujo (talk) 17:15, 9 March 2008 (UTC)[reply]
They could, if they wanted to. There are rainbow table sites running things like ophcrack and things of that nature. Needless to say, one should be cautious in using the same password for everything. --98.217.18.109 (talk) 20:08, 9 March 2008 (UTC)[reply]
The md5 keyspace has been 100% indexed. There are fast (<5 second) lookup bots floating around IRC :D\=< (talk) 18:54, 10 March 2008 (UTC)[reply]
But they don't have to hash it- fact is, they have your password in plain text during the signup script, and if they don't hash it and throw it away like they should, they can do anything they want with it. My friend, a beginner PHP programmer, didn't even think to use hashes and could see any of his users' passwords in phpmyadmin. I wrote my own login script and hash the passwords with md5, but in the user manager I link the hashes to their lookup query on rednoize and most of the test accounts' passwords are in the hash database. :D\=< (talk) 19:45, 9 March 2008 (UTC)[reply]
And of course there is the worst-case-scenario of an unscrupulous programmer using their programs to just harvest passwords (e.g. [2]). --98.217.18.109 (talk) 20:08, 9 March 2008 (UTC)[reply]
They probably have it in plain text every time you sign in. It would be trivial to write a script that emailed a programmer a users password before hashing it. You cannot trust others to safeguard your password just because they have a website. So, always assume that the site is going to sell your account info for profit and give them as little as possible. -- kainaw 01:19, 10 March 2008 (UTC)[reply]
This is why everything except website logins uses techniques such as challenge-handshake authentication protocol: they let you prove you know the password without it ever leaving your computer. --12.169.167.154 (talk) 10:30, 10 March 2008 (UTC)[reply]
I think to summarise the answer is that if an organisation does the "right thing" and stores hashes with salt values, peer reviews deployed code to make sure input values aren't intercepted, etc. you would be safe. There is nothing to guarantee that they do any of this, so you should assume that your password is not safe. -- Q Chris (talk) 11:16, 10 March 2008 (UTC)[reply]

A friend of mine (and one-time Wikipedia high mucky-muck of some description) once ran a site popular among a certain audience. The user table stored hashes (wired up to the real login system in the correct manner as described above), but *also* plain-text passwords purely for his own perusal, interest and amusement. If you give someone your password, you give them your password - even if they have software acting on their behalf. 81.187.153.189 (talk) 22:58, 10 March 2008 (UTC)[reply]

Any TLDs with very few second-level domains, and no barriers to registration? edit

Are there any two-character top-level domains that have very few domains under them, and that have no barriers to registration of such second-level domains (such as residency requirement, excessive cost)?

Reason for my question: I want have a very short and simple domain that can be typed quickly.--86.146.241.92 (talk) 18:40, 9 March 2008 (UTC)[reply]

Well, I suspect there's more reason to it than that, only because you're talking about a big-big savings of a single keystroke. --98.217.18.109 (talk) 19:42, 9 March 2008 (UTC)[reply]
Well, yes there are more reasons, not all of which I want to say (except to say there's nothing illegal or immoral involved). Anyway, hope someone can help...--86.146.241.92 (talk) 19:50, 9 March 2008 (UTC)[reply]
See Country_code_top-level_domain#Vanity_ccTLDs and the list below it- ccTLDs with asterisks are open to foreign registration. :D\=< (talk) 20:01, 9 March 2008 (UTC)[reply]

Free the Space edit

G'day... My XP laptop reports a different amount of free space available on my external hard-drive compared to the Mac at work. I deleted a whole lota files at work yesterday... for eg. its stating 600Mb free on my XP machine while on the Mac it says 11.5Gb is free for the same partition. I've emptied both recycle bin and trash can without any difference. Cheers for any advice :1 Boomshanka (talk) 20:53, 9 March 2008 (UTC)[reply]

Are you sure that your external drive has a recycle bin and a trash can? I thought external drives do not have recycle bins. Kushal 05:07, 10 March 2008 (UTC)[reply]

That depends on how Windows recognizes it. If in My Computer it shows as a hard disk, then it has a Recycle Bin (\Recycled) and (sometimes) a System Restore Information folder. If it shows as a removable disk, then Windows doesn't create such folders. --grawity talk / PGP 16:20, 10 March 2008 (UTC)[reply]
So that means XP and Mac OS report deleted files differently? ...and so one OS would not recognise the fact the other OS had marked them as deleted. That makes sense I guess; ok, thanks Boomshanka (talk) 21:54, 10 March 2008 (UTC)[reply]