Wikipedia:Reference desk/Archives/Computing/2008 July 2

Computing desk
< July 1	<< Jun | July | Aug >>	July 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.

July 2

What's that port?

I have 2 types of mystery port on the back of my IBC PC that I was wondering if someone could identify. They don't seem to be documented on Computer port (hardware), but I would like to get them added.

I thought these were serial ports, but the symbol is different than the standard one.

I have no idea what this is; I had never even noticed it there before.

Thanks in advance, Beland (talk) 02:17, 2 July 2008 (UTC)

I'm pretty sure that second one is a really old external SCSI port. The other two could be old serial ports. They are 9-pin D-shaped, perhaps just an old symbol. Useight (talk) 02:46, 2 July 2008 (UTC)

The second port could be where a docking station connects. Have you tried Googling the model of your PC to see what the web reveals?

Atlant (talk) 12:18, 2 July 2008 (UTC)

Looks like a 80-pin SCSI connector to me [1] JessicaN10248 16:04, 2 July 2008 (UTC)

The second one appears to be a centronics-style connector. If the computer is a desktop, it's either a printer port or a SCSI connector; if it's a laptop, it's probably a docking station connector. The first one is probably a pair of serial ports with a non-standard symbol. Given the symmetry of the symbol, it may be RS-422 rather than the more common RS-232. --Carnildo (talk) 21:53, 2 July 2008 (UTC)

The top ones are almost certainly serial ports. The bottom one I expect is also a SCSI port. It's not like any parallel port I've seen- the Centronics end is usually at the printer side (the computer is DB25), and it's way wider than a printer Centronics connector. JeremyMcCracken (talk) (contribs) 21:59, 2 July 2008 (UTC)

Adding entries to the registry using .reg files

I found a list of malicious web sites here: http://www.2-spyware.com/news/post334.html. I'm trying to add them all to Internet Explorer's restricted sites (Tools --> Internet Options --> Security --> Restricted sites) via the Registry. I got regedit to create the keys, but they have no values. I double-click on the .reg file I created in Notepad and I see a message box saying that it worked each time. Here's what I enter in the file:

Windows Registry Editor Version 5.00

[HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Internet Settings\ZoneMap\Domains\expertantivirus.com]
“*”=dword:00000004

Any help would be greatly appreciated.--Hello. I'm new here, but I'm sure I can help out. (talk) 04:33, 2 July 2008 (UTC)

The quotes should be like this " (hex 0x22, Alt+034). --grawity 12:52, 2 July 2008 (UTC)

Phew. Thanks, Grawity. It works like a charm.--Hello. I'm new here, but I'm sure I can help out. (talk) 13:59, 2 July 2008 (UTC)

FireFox 3

I currently run FireFox 3 on a two month old laptop running Windows Vista. Sometimes when I am using it, the program will freeze, and I cannot click on anything for a few seconds, I cannot scroll, and anything I am typing will stop until the freeze is over. Is this a bug with FF3? Or is it my internet or computer? Is there a particular way to diagnose what is wrong?

As a completely unrelated question, is there any significant security attained by using an https as opposed to http? Thanks. seresin ( ¡? ) 05:12, 2 July 2008 (UTC)

I don't have an answer for why Firefox 3 freezes, but I can tell you that generally, https is more secure that http when transmitting sensitive information over a public network (eg. Internet) because it encrypts your transmission. However, it won't help if your system is compromised, or the server you are talking to is. Rilak (talk) 06:00, 2 July 2008 (UTC)

That's happened to me on older versions of firefox. Ususally it was because it was loading some heavy java applet, flash program, or it was executing a particularly heavy piece of javascript. Does any of these sound right?

As for https, it's substantially more secure than http. Normally, every piece of data passes through something like a dozen to two dozen computers, each one of which can read everything you send and retrieve. Also, it's not authenticated, so you can't be sure that you're actually talking to who you think you're talking to. Your communcation can be changed in transit. None of that is possible with https. If a site offers it as an option, use it (for instance, if you use gmail, don't go to http://mail.google.com, go to https://mail.google.com ) 83.250.202.36 (talk) 13:42, 2 July 2008 (UTC)

I'd blame the freezing more on Vista. I don't know about FF3, but I have a laptop and desktop where FF2 work great, and my girlfriend has a Vista laptop where FF2 dies constantly (as well as some other programs). JeremyMcCracken (talk) (contribs) 22:01, 2 July 2008 (UTC)

I would have been inclined to agree, but I used FF2 on my old laptop with Vista, and never had this specific problem (I had other problems, but I think they had to do with my internet connection). seresin ( ¡? ) 22:49, 2 July 2008 (UTC)

The same thing happens to me often. It's generally because of—as someone said before—a heavy java applet, or possibly just a gigantic page. ¢rassic! (talk) 08:35, 6 July 2008 (UTC)

Java Swing: How to get pixel values from components?

I have a Java Swing component (a javax.swing.JPanel) that I want to draw stuff on and then change the colours of the pixels. But how can I get the colour value of a specific pixel? Neither javax.swing.JPanel or java.awt.Graphics seems to define any way to do it. Am I supposed to somehow read the contents of the component into a java.awt.image.BufferedImage and use its pixel-level methods? JIP | Talk 05:30, 2 July 2008 (UTC)

It's been a long while since I've done stuff like this, but as memory serves, the ImageConsumer interface is used for this, specifically PixelGrabber. I think you can only use this with an Image object directly (not a Graphics object), so I suspect you need to use dubble-buffering to use it. See also [2] 83.250.202.36 (talk) 10:16, 2 July 2008 (UTC)

You can use java.awt.Robot, although that will see any windows that are obscuring your components and can't usually be done from an applet (if that's relevant). --Tardis (talk) 15:48, 2 July 2008 (UTC)

No, I'm afraid this won't do. What I want to do, is I want to load an image into a JPanel (this I am already doing), and then I want to change the colours of some pixels depending on what colours they already have. This should be done dynamically, meaning that once I have changed the colours of some pixels, if I change them again, it's done by the colours they currently have, not by the ones they originally had. JIP | Talk 19:24, 2 July 2008 (UTC)

Here's what I'd do: load the image one time and use PixelGrabber to get the pixels (can I assume you are loading the image from a file? Either way, get it as an image, and use PixelGrabber). Save the int[]-array. Modify the captured pixels however you want. Use the pixels to create a MemoryImageSource, but remember to keep the original int[] array. Next time you want to modify the pixels, modify the int[]-array, and call MemoryImageSource.newPixels(). And then update the widget, or whatever. The javadocs for MemoryImageSource makes it quite obvious how this all works (read the intro). Will this work for you? Or am I misunderstanding what you want? --Oskar 18:59, 3 July 2008 (UTC)

Oh, yeah, I guess you have to set the MemoryImageSource to be animated. As I said (that was me, the first response, forgot to log in), I haven't used this kind of stuff in a long time, but I think this will work for you. Let us know if you need more help. --Oskar 19:02, 3 July 2008 (UTC)

Because I had absolutely nothing to do today, I put together a small little demonstration of what I think you want. --Oskar 22:10, 3 July 2008 (UTC)

video conferencing

what is video conferenciang?

How is it works?

Is there any advanced techniques used in this now? —Preceding unsigned comment added by Alex.snowfall (talk • contribs) 07:37, 2 July 2008 (UTC)

You should probably start by reading our article on videoconferencing and then come back if you have any questions left after doing so. -- Captain Disdain (talk) 08:55, 2 July 2008 (UTC)

There are some pretty advanced items in it. I work for a department that uses H.323 encoder boxes (not through a computer; it's a standalone unit). We operate multisite conferences for high schools to connect to the university for post-secondary classes. JeremyMcCracken (talk) (contribs) 22:04, 2 July 2008 (UTC)

Copying hard drive

I have a laptop and a portable hard drive. I want to use the Portable HDD to move files etc, and as a sort of backup. I have read the article on HD cloning. I do want a clone, i just dont really understand. Help; which software should i use. And how do i do it. Quidom (talk) 11:24, 2 July 2008 (UTC)

What size, in GB, is each internal and external HD (hard drive), and which operating system (eg. Windows XP, Mac, Linux) are you using? This is important because software available for one OS might not be available for another.78.149.63.101 (talk) 11:17, 2 July 2008 (UTC)

I have a laptop hardrive of around 120GB and portable of around 250GB its on windows XP, the portable is BUS powered and is a toughdrive Quidom (talk) 11:25, 2 July 2008 (UTC)

VGA connector pin 9

I have a vga connector that lacks pin 9, question what is "+5 V (DDC)" specifically what is DDC, is it Display Data Channel?

When this plug is used to connect a LCD monitor it seems to function correctly - what can possibly go wrong??87.102.86.73 (talk) 11:54, 2 July 2008 (UTC)

Pin 9 is +5V DC (direct current), not +5V DDC. I edited the article. It appears that since it is a DDC signal on the cable, someone assumed "DC" was a typo for "DDC". It is very possible that the monitor/projector does not need the 5V reference. If you plug this into a monitor/projector that requires the 5V reference, the monitor/projector will be looking at at the cable and seeing that a positive signal is around 0V (no connection) and a negative signal is 0V (the ground connector). It will then have difficulty telling positive and negative signals apart. -- kainaw™ 12:25, 2 July 2008 (UTC)

Misses pin 9.

Interestingly enough, the reference image at the article is missing pin 9 too! (See image). User:Krator (t c) 12:35, 2 July 2008 (UTC)

Which curiously looks a lot like (identical except colour) my plug...

Question - anyone know why this pin is 'often' ommited? and why is PIN 4 there when it's marked 'not connected'... (and thanks so far)87.102.86.73 (talk) 12:51, 2 July 2008 (UTC)

Actually, the "+5V DDC" labelling was (somewhat) correct. Pins 9, 12, 14 and 15 all had new functions to support the DDC Specification. In the original Spec pin 9 provides no function, hence it's removal, and it is also only Optional in DDC1. If i can find official documentation supporting this i will post it. - Jimmi Hugh (talk) 14:20, 2 July 2008 (UTC)

If a pin that provides +5VDC is missing, the sensed input depends on how the input is wired. If it has a pullup resistor, it will go to +5VDC; if a pulldown resistor, it will go to 0VDC; if no resistor, it may ambiguously float. --—— Gadget850 (Ed) ^talk - 16:12, 2 July 2008 (UTC)

filename.ext in Windows

(Inscrutable late-noughties proprietary OS blues 1)

The Explorer that comes as part of Vista hides the filename extension, so that mystupidface.jpg appears as mystupidface -- which might instead mean mystupidface.JPG or mystupidface.jpeg or indeed mystupidface.png or whatever. More oddly, I can't find an settings menu to change this. How do I get Explorer to reveal all? -- Hoary (talk) 14:41, 2 July 2008 (UTC)

You can't fine the settings menu? Or the setting? The menu can be found either through the Control Panel, "Folder Options" in Classic View (Can be selected at the side), or through the Tools Menu (Use the Alt key to bring up hidden menus) and choose "Folder Options..." from the "Tools" menu. The actual option is on the "View" tab, as "Hide extensions for known file types". - Jimmi Hugh (talk) 14:53, 2 July 2008 (UTC)

Aha. I was looking for a setup menu within Explorer. Thanks, I'll make this change when I next use the Vista machine. -- Hoary (talk) 15:03, 2 July 2008 (UTC)

Just FYI, this behavior hasn't changed much since at least Windows 2000. And I'm of the opinion that no sensible person operates with "well known extensions" hidden; sooner or later, hiding those extensions will bite you.

Atlant (talk) 22:16, 2 July 2008 (UTC)

Extraordinary. At home, I still use a Windows machine, because I set it up in 2000 or so and it's been chugging along uncomplainingly ever since, because it's seldom connected to the internet, and because Win2k actually works rather well after it's been adjusted and supplemented in various ways. But all the adjusting and supplementing was so long ago that it seems I've forgotten a lot of it. I'd completely forgotten that this machine ever obscured filename extensions. The risks are obvious but more than that there are various irritations; I don't know what Microsoft could have had in mind when they provided this misfeature. -- Hoary (talk) 22:51, 2 July 2008 (UTC)

You'd be suprised how many people don't care about file types, and infact find the extensions more confusing then useful, especially when renaming files. I'm pretty sure, though it was a while since i last reinstalled a Vista Machine, that Proffesional (Or perhaps Ultimate) edition turns it off by default. Also, for the record, I still use Windows 98 and it also provides the feature. I'm not sure about Window 95, as i think it may have been added with IE4, back when IE versions were just as important in working out our feature set. - Jimmi Hugh (talk) 23:01, 2 July 2008 (UTC)

escape from Iphoto

(Inscrutable late-noughties proprietary OS blues 2)

Somebody's done something to the Mac OS 10.5 machine such that when I attach a CompactFlash card to a USB port, Iphoto starts up to display the content. Iphoto is very pretty, but I don't particularly want to use it. Finder/Preview are fine for my purpose. More bizarrely, when I drag a minimized image out of the Iphoto window and release the mouse button in order to copy it, the image merely bounces back and isn't copied anywhere. How can I copy out of Iphoto? -- Hoary (talk) 14:41, 2 July 2008 (UTC)

This is one of the few places where Mac OS X is definitely not straightforward. The preference you're looking for is in the Image Capture program (in your "Applications" folder). One of the "General" preferences is "When a camera is connected, open..."; select "No application" (or Photoshop or whatever).

Atlant (talk) 22:13, 2 July 2008 (UTC)

Splendid. I'll make that change a couple of hours from now when I get to that computer. Then I'll no longer have to use the Wintel across the room merely in order to get images off the card. -- Hoary (talk) 22:53, 2 July 2008 (UTC)

Stupid question about smart antenna

Does the CEA-909 standard, in addition to physically or electronically rotating the antenna based on channel, also electronically tune to a particular channel and filter out the rest, similar to how a Channel Master jointenna is tuned ? StuRat (talk) 14:48, 2 July 2008 (UTC)

My best guess so far is it might be possible if the antenna implements it, but smart antennas will likely focus on directional control. Since the CEA-909 standard costs money, here's what I've based my guess on:

• From the CEA-909 article, it says 14 bits of information are sent to the antenna. It would be possible to send RF channel information in 7 bits (there would need to be 68 possible values, to represent RF channels 2 to 69), in which case the other 7 bits might represent direction or other information. So based on the number of bits, it's possible channel information could be sent to the antenna.

• One of the references listed in the CEA-909 article reveals just the table of contents of the CEA-909-A standard: ANSI/CEA-909-A Preview. In that table of contents, several example applications are mentioned, including "Mast-Mounted Preamp Seeking Gain Control Bits Plus RF Channel Number". That further suggests it's possible to send the RF channel number to the antenna, but depending on the type of antenna, it may not do anything with it.

• I found an ATSC newsletter with a short description of how the smart antenna system works: The Standard, April 2003, page 2 and 3. I also found another description from BroadcastEngineering.com: Smart antennas. And here's a manual for a Funi/Sylvania antenna: Smart Antenna DTA-5000 Instruction Manual. All of these sources describe antennas that use direction and gain settings, suggesting those will be the most common parameters that smart antennas use. The BroadcastEngineering.com article also says "In addition to selecting different azimuth directions, units can operate with different levels of RF amplification." Is that the type of channel filtering you were asking about? --Bavi H (talk) 16:40, 4 July 2008 (UTC)

Well, I've got a problem with two strong adjacent UHF TV channels (31 & 33) interfering with the weak channel I want (32), so a directional antenna will help, but I'd also like to block the adjacent channels, and was wondering if any smart antenna, now or future, would be likely to solve my problem, or if I'd also need a Channel Master jointenna to block the stronger adjacent channels. StuRat (talk) 01:20, 6 July 2008 (UTC)

security in the properties window

What is "security" in the properties window on Windows XP? It says "this file came from another computer and is blocked to protect this computer". What does it mean. You can see what I'm talking about here: http://img381.imageshack.us/img381/1781/whatthefrelldoesthismeami1.jpg FasterMaster (talk) 15:12, 2 July 2008 (UTC)

It means Windows will warn you before opening it, until you uncheck that. I'm not sure what marks it as being from another computer. JeremyMcCracken (talk) (contribs) 22:09, 2 July 2008 (UTC)

This is part of Windows Attachment Manager. Certain file types downloaded from untrusted zones, attachments saved from Outlook Express, or files copied from other computers may have this flag set. See "Description of how the Attachment Manager works in Windows XP Service Pack 2" or google search "this file came from another computer and might be blocked to protect this computer". -- Tcncv (talk) 23:12, 2 July 2008 (UTC)

Sharing a monitor but not its display

Related to the mental poker question above: how might n people (we can start with $n=2$) using one computer (and one monitor) play a card game with secrecy? Certain physical tricks could be contrived, like a partition attached to the front of the monitor to restrict sightlines, but those are inconvenient and do not extend well to $n>2$. Displaying one person's information at a time, and showing nothing at all while another player moves to the computer, is also prohibitively inconvenient, especially because it is useful to be able to see one's cards continuously rather than only when it is time to select one. One idea that may be useful is to have some sort of multiphase approach: the players could be shown their cards individually, and then they could note them offline (not by finding the same cards in a shared normal deck for the game, though, as the missing cards would reveal the other players' hands) for later use. Obviously, though, an interface that was more than type the card from your paper that you wish to play: would be more interesting and convenient.

Card games typically have very little information to display: Rook's important and visible state can be represented by a short list of colored labels like RK 7 9 13 8 9 10 11 14. So I imagine that, perhaps with suitable obfuscation or additional confusing information (whose nature would answer the question), several such displays could share a monitor. However, the requirement that the secrecy survive the interaction with the game and the acquisition and revelation (play) of individual cards prevents the use of a simple substitution cipher to encode the hands, and the requirement that humans still be able to play the game quickly prevents the use of complicated encodings that could trivially provide the secrecy. Any ideas? --Tardis (talk) 15:36, 2 July 2008 (UTC)

Perhaps using 3-d movie technology? --Sean 18:09, 2 July 2008 (UTC)

Yes, If number of players == 2 then you could use 3d technology like Sean suggested. This monitor apparently puts out two separate images polarized differently. Give each player one polarized lens. But how to stop them from simply rotating the lens by 90 degrees? APL (talk) 20:12, 2 July 2008 (UTC)

Three suggestions :

• Colored filters. Draw the cards faintly in a certain color over a very noisy background in two other colors. Player looks through the appropriate filter to see the cards. Drawbacks : could really only work for three or less players, and it is possible to figure those things out of you stare at them hard enough.
• Shutter glasses. Usually used for stereo, these could be used to show each player alternate frames. (Noise could be introduced over opponents' cards to minimize the effects of ghosting.) Drawbacks : would require custom hardware that was more expensive than simply providing extra screens. Would require very high refresh rate for more than two players, otherwise the shutter speed would become low enough to give the users a headache.
• Cypher. Print out three lists of One time pads at the beginning of the game, players then use these to decrypt their cards. Drawbacks : The most obvious, easy to use cyphers would not properly hide flushes and four of a kinds. A lot of mental overhead for the players. Relies on on-paper OTP.

If I was making this system, I'd try the colored filters. However, it is probably not secure enough if money is changing hands. APL (talk) 20:12, 2 July 2008 (UTC)

Taking turns in front of the computer is cumbersome, but it can be made easier if you avoid the need for the players to go to another room or behind the display between turns. One approach is to show the information in a really small font and a low-contrast colour scheme so that you have to be very close to the screen to read it. You might also get the kind of directional filter they use at banks so that a player can't see anything unless they are directly in front of the screen. The best solution would be if the computer is a hand-held (or a very light-weight laptop) so that you can simply pass it around from player to player, but I suppose you don't have that kind of a luxury. 84.239.133.47 (talk) 07:42, 6 July 2008 (UTC)

I think we're dealing with a fundamental problem. If there are n viewers, and each one needs to see private information from the same source, there are only a few options:

1. Encryption of some sort. Each viewer can only decrypt his/her own channel. This was addressed above (and I think it's the best idea).
2. Polarization. This is related to the technology of the display. I like it but of course cheating is trivial. One doesn't even have to rotate the glasses - simply take them off (unless there is some sort of interference channel to force the glasses to be on). Monitors can't do polarization regardless.
3. Switching views between users. This really is the correct solution for normal hardware. Each player could have a secret keystroke identifying him/her to prove that the other person isn't looking.
4. Physical barrier. This isn't a solution, just a cheap way of multiplying the displays.

The secret is ... what can a user do to get information securely.

1. How about image ciphers or something like that? If the user starts a session with a set of fake cards (say a dog equals a heart, and any number is 2 less than it really is), it could theoretically work. However, this system would be very easy to break I think.
2. Trust. Have the users vow not to read the other side. Then give them the polarized glasses above.

I think if you post to 70s and 80s video game newsgroups you might find a better answer than here. —Preceding unsigned comment added by Adamnelson (talk • contribs) 01:06, 8 July 2008 (UTC)

Google definitions

As many of you no doubt know, if you enter a search term in Google with "define:" in front of it, Google will give you definitions from various web pages (including Wikipedia). Here is an example.

I will be helping a professional organization put together a list of definitions to key terms used within the academic field the organization represents. The definitions themselves are to be written by members and peer reviewed.

Is there any way to optimize my writing of this so it shows up in Google's "definitions"? I haven't been able to find anything about how Google chooses which sites to use. (searching for "Google definitions" comes up with, predictably, definitions of the word "Google"). Any thoughts? I've looked at some of the other pages that were featured and there's no common thread that I can see, other than the fact that the word to be defined is usually in bold and followed by the definition... --98.217.8.46 (talk) 15:36, 2 July 2008 (UTC)

Thought i've never managed to cause a definition listing on Google, the following article has apparently been useful to other's attempting it [3]. Mostly it looks at how existing sites manage it, and i believe it draws advice from a related patent application. As long as you structure the website containing definitions in a way that is easy to interpret, and follow some of the advice given, it seems that Google's algorithms for such things are quite thorough. - Jimmi Hugh (talk) 15:52, 2 July 2008 (UTC)

OK, that is helpful. It looks sort of like I thought -- having a domain name that looks "authoritative" in some way (in this case, a high-profile EDU will help), and structuring it in the bold word + normal definition, all in neat paragraphs, seems to work out fine. Well, we'll see! --98.217.8.46 (talk) 16:02, 2 July 2008 (UTC)

Purely speculative, but I'd guess that sticking it in a dl would be your best bet - after all, it's for definitions by, well, definition. — Matt Eason ^{(Talk • Contribs)} 16:38, 2 July 2008 (UTC)

Domain squatting

Is there an estimate to how many registered domain names are just "squatting" pages? (Either no page at all, or those "parked domain" pages) It seems rather ridiculous to me to have a system that so easily allows itself to be co-opted for pure speculation (registering any name or word you can think of for a few dollars and then hoping against hope that someone will want to pay you big money for it someday), but it'd be nice to have actual numbers. --98.217.8.46 (talk) 15:50, 2 July 2008 (UTC)

I do not know the number, but I want to point out that domain squatting is not just holding onto a domain name and hoping someone will pay for it someday. Many squatters register domain names as soon as they expire and demand thousands of dollars to give the domain back to the company that was using it before it expired. Worse, a squatter in New Orleans used the extreme lack of security with GoDaddy to literally steal two of clients' domain names when they weren't expired. GoDaddy required a minimum of $1,000 to research and correct the action. The squatter asked for $900 to return to the domain name. -- kainaw™ 16:43, 2 July 2008 (UTC)

Well, I consider that something more than just "squatting" (that's just plain old extortion), but yeah. I'm really just interested in the numbers on squatting. --Fastfission (talk) 19:41, 2 July 2008 (UTC)

That's not really "squatting" so much as "stealing" or "exploiting". --98.217.8.46 (talk) 19:47, 2 July 2008 (UTC)

You can also register a domain for a trial period (if I remember correctly, it is 5 days). During that trial period you can setup your link farm page and wait to see if there is enough interest in the domain to justify keeping it.--droptone (talk) 12:10, 3 July 2008 (UTC)

CGI

I've read that CGI is scripting used to interact with a server,^[4] but I've also read on Wikipedia that it's a protocol to interact with a server.^[5] An instructor in a video tutorial the other day said that the second part of URLs like this are CGI:

http ://en.wikipedia.org/w/index.php?title=Wikipedia:Reference_desk/Computing&action=edit&section=new

But wouldn't the URL be HTTP? Also, it appears that files with the extension .cgi are in Perl and the /cgi-bin directory is used mostly for Perl scripts. Could it contain PHP? Are all PHPs and JSPs part of a CGI?--Cgi-bud (talk) 17:28, 2 July 2008 (UTC)

The CGI page explains a little about it. It is a protocol that allows a server to execute anything that can be interpreted or is an executable. For example, on a webserver, you could have a page that calls a file "process" (note the lack of extension) that is an executable file that was written in C. The CGI protocol defines how arguments are passed to the program. So, if the request (assuming a web server) is a get request, the arguments are available in the environment variable 'QUERY_STRING'. If it is a post request, the message body of the request is available on STDIN. The program then executes like a normal C program, using STDOUT to pass the output to the web server. It is called Common Gateway Interface because it handles everything in the same way so that many different types of scripting/programming languages can be used. This includes PHP, PERL, Ruby, and any other interpretted language or executable. JSP is a little different. Take a look at the JSP page for more information. Leeboyge (talk) 18:06, 2 July 2008 (UTC)

And I think you're right that the variables are being passed using HTTP form standards (in this case, GET, but it could also be POST). But the CGI can then take those variables and use them. --98.217.8.46 (talk) 20:12, 2 July 2008 (UTC)

I see. That makes sense. Thanks for the replies.--Cgi-bud (talk) 02:08, 3 July 2008 (UTC)

Just a little note, everything after ? in the URL above is part of the querystring (GET), as is said above. While CGI can access those variables, there is nothing 'CGI' about the URL or querystring.

Memory compiler

What is a memory compiler (SRAM compiler, RAM compiler, EEPROM compiler etc.), for example as mentioned here? Thanks, --Abdull (talk) 18:04, 2 July 2008 (UTC)

Firstly, absolutely fantastic question, i've never before found such a lack of information on a topic mentioned in so many places. I had quite the time researching it and am still left with numerous questions that i've actually now passed onto a friend at ARM who will ask around the people in the department, as well as other places. Firstly, i found the following two articles which provide pretty much the best descriptions i could find [6] [7], however i'm sure someone might be able to provide better sources. Though i haven't had a thorough study of either, they don't seem to explain if it is a Compiler in a traditional Text->Text sense, application software that turns a description into a layout, or as is implied in the second article, is it simply a class of specifications for an interface that provides flexible memory on demand (or at least in a manner making it changeable at times). However, the purpose seems somewhat more apparent. It is a "system" where by details of the interaface between a computer system and memory communicate. The details passed to the compiler, include details such as word sizes etc, as well as low level details I am afraid i don't know anything about. - Jimmi Hugh (talk) 19:40, 2 July 2008 (UTC)

It seems to be some sort of design program that will generate memory banks of different types for your chip design - in other words say you are designing an ARM chip and want 16kB of cache built in.. and your chip has a space of say x by y mm spare to put that cache in.. the 'memory compiler' will design your cache so that it fits in the space on the chip, with the correct buses etc...87.102.86.73 (talk) 21:12, 2 July 2008 (UTC)

In other words it 'compiles' a design for your custom silicon memory parts...87.102.86.73 (talk) 21:13, 2 July 2008 (UTC)

Already at this point I thank both of you for your comments. The IC industry seems to throw out terms that defy authoritative definition. Jimmy, I'm interested in what your friend at ARM will have as a definition for memory compiler. Maybe you are interested in creating this article then? At 87.102...: so a memory compiler turns a specification for a piece of on-chip memory into a part of the whole Integrated Circuit Layout? Actually I'm surprised that on-chip memory is (always?) compiled. I'd have expected memory being added to an IC by use of a standard cell library that (so to see) offers "precompiled" components. --Abdull (talk) 21:48, 2 July 2008 (UTC)

(I guessed from the available info. 'memory compiler' is a new term to me)

Yes. I'd imagine that it would produce 'bitmaps' (or whatever they use) to produce the masks, as such it would most likely utilise a standard cell library to form the invidiual logic gates etc. That was my interpretation of the ARM page, I think I was close, but not necessarily spot on.87.102.86.73 (talk) 22:15, 2 July 2008 (UTC)

Well, i didn't get much more information from ARM than what has been said. However, apparently, the term Memory Compiler isn't constant, with a few definitions, all relating to the creation of the interface with the memory or the memory itself. In ARM's case it is a product containing software for helping the Electronic design automation process, by generating elements of the design based on your specifications. The specifics are seemingly apparent when you use the software, if you already know the details of designing memory. I haven't run the software yet, but i may get a copy if this still isn't enough, or in order to ensure an article on the matter is created. - Jimmi Hugh (talk) 10:16, 7 July 2008 (UTC)

Videos on iPod

Does anyone know how many Megabytes a 25 minute video takes up on an iPod? I want to download an episode of Family Guy from iTunes, but I have a 4 GB Nano and I don't want to download the episode if it takes up to much space. Thanks! Grango242 (talk) 20:10, 2 July 2008 (UTC)

Depends on the quality of the video 4.158.219.215 (talk) 04:15, 3 July 2008 (UTC)

Well thanks anyway, but I'm glad that I asked at the miscellaneous desk too. Grango242 (talk) 01:10, 4 July 2008 (UTC)

?Query??

Message of peace to all!I'm working on developing a network for a company ,just a student level work.Tell me please if i have 5 buildings in a company 50m apart from each other, 22 computers in each building 20in first floor and 2 in 2nd floor then where should be the location of switch.Why usually star topology is preferred over other topologies.--Catherine.michi (talk) 20:56, 2 July 2008 (UTC)

Star topology redirects to star network, which should answer your question better than any short answer here could. If you need help with topics covered in the article, please ask. -- kainaw™ 00:59, 3 July 2008 (UTC)