Computing desk
< August 8	<< Jul \| August \| Sep >>	August 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.

August 9

Mobile phone apps

Hi. I have a few questions that I'm hoping someone can help with. I am mostly interested in the situation in the UK but any general info would be welcomed too. Assume complete ignorance on my part.

1. Do mobile phone "apps" that need to communicate with the outside world (e.g. weather forecast, news, social networking, etc.) all use the Internet, or do they also utilise the traditional phone network? Or is it now the case that there really isn't any difference?

2. Such "apps" appear to often provide functionality that on a PC would be accessed via a web browser, so that most people do not need to have dozens of small specialised pieces of software on their PCs with each one accessing a different service. Why is the mobile phone model different? Is it anything to do with the difficulty of implementing a general-purpose browser for such a small screen?

3. If a charge is made, then how are these services typically charged for? Are there mechanisms whereby access to a particular Internet service is automatically charged to one's phone bill, like when one makes a premium-rate phone call? Or are payments made separately using the same methods as one would use with PC-based access to such services?

TIA... 86.135.171.33 (talk) 01:02, 9 August 2010 (UTC)[reply]

Well: 1:These apps will generally use the Internet, not the phone network. I'd imagine that the phone network cannot send information because of lack of bandwith. 2: I don't think so, it's called giving people more functionality and raising brand power. Also they can put ads on their apps. 3: That would depend on the moblie provider.Sir Stupidity (talk) 03:09, 9 August 2010 (UTC)[reply]

Thanks. Regarding 3, do you know for sure that there definitely are some mobile providers that use this kind of charging system (even if others don't), or are you just speculating that some might? 86.135.171.33 (talk) 03:14, 9 August 2010 (UTC).[reply]

Yes, they use the mobile phone network. How would they connect to the internet without it? (unless you're using WiFi at home or a coffee shop). --mboverload @ 04:16, 9 August 2010 (UTC)[reply]

1-They use the mobile phone network to connect to the internet.

2-Purpose-built native apps run much faster on a low-powered device like a phone. Apple tried to make developers build apps via the web but they sucked in general.

3-A charge is made to your iTunes store account (various payment methods). I don't know how Android (Google) phones work with their store. --mboverload @ 04:33, 9 August 2010 (UTC)[reply]

When you ask about "the traditional phone network" I assume that you were not thinking of landlines (the most traditional of phone connections). Yes, mboverload's answers are correct, of course, but Sir Stupidity is correct in stating that there are bandwidth issues in accessing the internet over mobile networks, and this is a serious problem with mobile devices in some rural areas. Dbfirs 07:26, 9 August 2010 (UTC)[reply]

On the question of how the apps get their data - neither Sir Stupidity nor mboverload's answers are quite accurate. The "traditional" mobile phone network (which is still generally used for mobile calls, texting, etc., has a very low bandwidth and would not be much use for the apps. Mobile phone operators have built higher bandwidth data networks (either 3G or HSDPA) to carry data over the air interface. The data is then carried over an upgraded "phone network" before typically being forwarded to the internet. To summarise - mobile data is carried over upgraded mobile data networks and then the internet. --Phil Holmes (talk) 07:48, 9 August 2010 (UTC)[reply]

Yes, I wondered if I should mention this, but in some areas there seems to be slow internet access in places where the companies claim that they have not upgraded. Is internet data impossible over an old-fashioned plain mobile phone network? Dbfirs 12:02, 9 August 2010 (UTC)[reply]

It is possible on 2+G networks, using for example GPRS or EDGE. Let's not forget the first iPhone didn't even have 3G (despite it being available in most smart phones of the era). Note that depending on the network and phone, 3G may be used for the voice and SMS where available, it will definitely usually be used for MMS. From the apps POV, where it uses GPRS/EDGE or UMTS/HSDPA should usually be transparent, it will use whatever the phone is connected with and the phone will usually connect with the highest speed network available, falling back as necessary, unless of course you manually disable one for whatever reason (I know someone with a 3G phone who didn't use WAP or the internet and disabled 3G because they found it reduced battery life).

Note that many countries still lack widespread 3G networks, some of them may not have them at all. The iPhone 3G & 3GS also were only tri-band so may not be able to use the 3G network even if it is available in some countries, in NZ for example the iPhone is sold by Vodafone who use 900mhz for much of their coverage outside major urban locations. In fact, we are somewhat of a rarity in having 900mhz and 850mhz used for differing 3G network I believe, it's not uncommon in Europe that only 900mhz is used.

I used to use the internet on my Panasonic VS2, one of the key things stopping me was the price not the speed, this was GPRS only (even if it was EDGE, Vodafone NZ doesn't support that). If you are downloading MP3s or regularly browsing the internet or visiting the reference desk or whatever then GPRS and probably even EDGE is going to be painfully slow, but for just visiting news sites and stuff, it's IMHO acceptable particularly if you use wap sites or proxies which reduce the images and stuff. Similarly when it comes to apps, it depends on what they do. If they just submit high scores online then I'm sure you'll be fine with GPRS or EDGE, if they upload high res photos every minutes, then that'll likely be unbearable.

Nil Einne (talk) 12:08, 9 August 2010 (UTC)[reply]

I was simplifying somewhat, and so didn't mention GPRS. Do note, however, that even if GPRS is used as the air interface, the data is generally groomed off the "phone network" and onto the internet by a GGSN. --Phil Holmes (talk) 16:02, 9 August 2010 (UTC)[reply]

Thanks for all the info guys. A follow-up question if I may. "A charge is made to your iTunes store account (various payment methods)." (from Mboverload, above) Does this mean for an initial app download, or could that be on a per-use basis too? I would be interested to know more about the possible charging methods used for mobile phone Internet services (I mean services such as news, weather, financial data, etc. etc., not for the Internet connectivity per se) and how prevalent the various charging models are. Anyone knowledgable about this? 86.135.24.96 (talk) 18:21, 9 August 2010 (UTC)[reply]

Regarding the app model, perhaps it would make a bit more sense if you look through the other end of the pipe, i.e. at your PC from the point of view of the phone. What are the "apps" on your PC? A word processor, a few games, and this magical super-app called a web browser that can handle "whole bunches of stuff". 'zat help? DaHorsesMouth (talk) 00:36, 10 August 2010 (UTC)[reply]

merging one image from multiple video frames

i've been doing a little experimenting on merging multiple similar frames into a video to make a better quality image, and i've seen something before on how to increase the resolution by averaging pixels or something like that, but i can't remember where it was. i've found various success with reducing noise and getting some more natural color off of dark video, but i'm just looking for any tips. i'm using gimp. thanks! 210.254.117.185 (talk) 03:42, 9 August 2010 (UTC)[reply]

And is there anything currently available that can approach the methods described here: [1] ? 210.254.117.185 (talk) 04:19, 9 August 2010 (UTC)[reply]

Pixel art scaling algorithms and Image scaling might be of use? --Mr.98 (talk) 14:52, 9 August 2010 (UTC)[reply]

What is the source of your video? Is it an interlaced video format, like NTSC or 1080i, so the point is to try to get more image data (albeit blurry) before getting to work sharpening the image? Comet Tuttle (talk) 16:28, 9 August 2010 (UTC)[reply]

There are several freeware programs used by amatuer astonomers to combine many images into one. I cannot recall any names or links. Do not know if they work with video, but at worst you could turn the video into still frames, which some free video players such as Gom player will do, and combine those. Some Google searching brings up for example Registax, Deep Sky Stacker, Drizzle (image processing), Registar. See also Super-resolution, in particular the external links about video super-resolution. 92.29.121.47 (talk) 10:55, 10 August 2010 (UTC)[reply]

Is it called a command-line interpreter, a command-line interface, a terminal editor or a shell?

The article on cmd.exe calls it a command-line interpreter. The article on Terminal.app calls it a terminal emulator. Windows PowerShell's first sentence has a link to command-line interface and shell (computing). The inconsistency is confusing me. Is there no single united term for these things? Should all these articles be merged or reorganized in some way?

No. They do not mean the same things. --85.77.181.105 (talk) 06:00, 9 August 2010 (UTC)[reply]

If the distinction isn't clear from reading the topics, then they're in need of work TEDickey (talk) 08:20, 9 August 2010 (UTC)[reply]

If I have this right, a terminal emulator can remote into a command-line interpreter, which uses a command-line interface. Unless the terminal emulator is a graphical one. From a Windows viewpoint anyway. Feel free to hack away at this ;) --mboverload @ 08:24, 9 August 2010 (UTC)[reply]

That sounds backwards. A terminal emulator can "contain" a "command-line interpreter". But command-line interpreters need not reside within a terminal emulator, since their functionality is much less than a terminal emulator. Even the Windows command-line interpreter uses only a small fraction of the functionality of the Windows command-window TEDickey (talk) 08:30, 9 August 2010 (UTC)[reply]

I'll try to provide a quick summary, so some stuff will be left out, but hopefully not too much.

What a shell does is read commands and interpret them. Some of these commands will consume input and/or produce output. This input and output (and also the commands themselves) will be in the form of streams of characters (in ASCII or some similar character set). Sometimes, commands are read from a file on disk (that's called a shell script), and input and output can also be redirected to files on disk, in which case no terminal is necessary. When the shell is used interactively, the character streams (commands, input, output) have to be connected to something that the user can see. That's where a terminal comes in.

What a terminal does is translate keypresses into characters for input, and characters into visible letters, numbers, and punctuation marks for output. The terminal was classically an external device with a keyboard and screen (or, even farther back, a keyboard and printer). The computer sends ASCII character codes through a cable to the terminal, and the terminal displays the corresponding glyphs on the screen. When the user presses a key, the terminal sends the corresponding ASCII code to the computer.

When a terminal and shell are used together in the usual way, the operating system sets things up so that the shell's output goes to the terminal, and the shell's input comes from the terminal. Other programs launched by the shell inherit its connection to the terminal, allowing them to do input/output to the terminal as well.

Nowadays, the classical terminals are all dead or in museums, but we have terminal emulators to do the same job: translating keypresses into ASCII input, and translating ASCII output into visible glyphs.

Since the terminal is no longer an external device, the connection is no longer a cable. The connection between terminal emulator and shell can be handled in two different ways. Option 1: the operating system provides an inter-process communication channel between the two programs which allows the shell to communicate with the terminal emulator using the exact same methods it would use to communicate with an external terminal. That operating system service is called a pseudo-tty, and makes it possible to mix and match shells and terminals and terminal emulators. Option 2: cram the shell and terminal emulator into a single inseparable unit. Guess which one Microsoft chose.

I've already covered shells used with terminals and shells used without terminals, so I might as well add the missing piece: a terminal can be used without a shell too. For example, the unix command xterm -e "w3m http://en.wikipedia.org/" starts up a terminal emulator and runs w3m inside. You're using a terminal emulator, the wikipedia main page is being displayed in it and your keypresses are recognized (for moving around and following links), but there's no shell involved.

The default behavior of most terminal emulators is to start a shell, because you have to have someplace for the terminal's input and output to go, otherwise you are effectively emulating a physical terminal with its data cable unplugged. Not very useful. Since something needs to be there, and shells are pretty useful, starting up with a shell is a reasonable default. 98.226.122.10 (talk) 23:48, 9 August 2010 (UTC)[reply]

98, the shell and terminal emulator are not a "single inseparable unit" in Windows. The terminal emulator is csrss.exe. Since it's always running and manages all console windows in a single process, it's less visible than a separate terminal process launched for each shell process, but it's roughly the same idea. X has a similar client-server split design with just one server for all running programs. If you added text terminal emulation to the protocols supported by the X server, that would be analogous to what Windows does. -- BenRG (talk) 08:20, 10 August 2010 (UTC)[reply]

So in unix, you implicitly create a shell by explicitly launching a terminal emulator, and in Microsoftland it goes exactly the opposite way. No wonder the poor users get confused.

By the way, none of this explains why command-line interpreter, command-line interface, and shell (computing) should be separate articles. I find no useful distinction between them. 98.226.122.10 (talk) 09:17, 10 August 2010 (UTC)[reply]

First of all it would be non-sensical to call python or ftp or R command line interfaces. Now they all have command-line interpreter modes, but they are not shells, unlike sh or cmd.exe. --85.77.213.188 (talk) 11:41, 10 August 2010 (UTC)[reply]

Agreed. Shells, interpreters, and command-line interfaces are are different concepts, though they appear superficially similar.

Command Line Interface: The easiest to explain is "command line interface." Anything that you type text-commands into, actuating particular functions, constitutes a command-line interface. For example, even the video game CounterStrike had a command-line interface. It would be invalid to call the game's internal console a "shell" though.
Shell: A shell is a specific type of command-line interpreter that resides between a terminal and an operating system. It has certain keywords or built-in functions that allow commands to define inputs and outputs, and actuate operating-system features. A shell can also request the operating system to start and stop other programs. There is a well-known, documented official standard published by IEEE, called "POSIX", which defines certain "must-have" features for any program to be called a "shell" (or at least a "POSIX-compliant" shell); so sh, bash, ksh, dash, and numerous other programs are POSIX shells, with extra features. Windows does not even try to be POSIX-compliant, so they have their own nomenclature, "command interpreter."
Terminal: A terminal is a special piece of hardware that displays output to a user, and receives input from a user. A software terminal, or terminal emulator, is a graphical software tool that does the same job, but is managed by an operating system's graphical subsystem (hence it is an emulated character-only device). It almost always goes to a "real" terminal (i.e. is rendered to a video output screen and receives input from a keyboard). But it could just as well be sent over a network; if a human is involved, at some point it will go to a hardware terminal somewhere, but the "emulation" means that this relationship can be more abstract than just a direct wire to a computer monitor. I have not seen the phrase "terminal editor" very much, but I assume that this refers to text-mode-only text editors, like vim.

These distinctions should be clear from the articles. We should review them to make sure it is obvious. Nimur (talk) 18:45, 10 August 2010 (UTC)[reply]

You didn't define command-line interpreter. When it discusses the history, that article lists various shells (and, uh, command-line operating-system interfaces). Surely, though, anything which has a command line interface has a command line interpreter behind that interface; so the article might as well list things like text adventure games as part of the history - except that in these situations the interpreter is inseparable from the interface (no pipes sending commands to the interpreter from text files). 213.122.22.8 (talk) 00:00, 11 August 2010 (UTC)[reply]

Windows 7 Administrator Problem

Hi, I was recently getting annoyed with my flatmate installing random programs he found on the Internet on the computer we both use for work and such. So, I made a new account that was an Admin, then proceeded to make the other accounts standard users. Then, I made the silly mistake of editing the Registry to make the other account invisible on the logon screen. Now, when a UAC dialogue pops up, I can't click "Yes", or enter an Admin Username or Password. This means I can't edit the registry to fix the original problem! Help!

Thanks, 110.175.208.144 (talk) 13:48, 9 August 2010 (UTC)[reply]

What happens at the login screen if you press ctrl-alt-del twice (make sure you do this without anyone being logged in)? I don't know about Win 7, but on XP you get the chance to enter a username and password. Try entering your administrator username and password using this method. You should then undo the registry tweak you used to hide the admin user.

Windows machines already have a built-in, hidden administrator account called "Administrator". If you change the password for that account you should be able to log into it using the method I mentioned above and it won't appear on the login screen. If all that work, you can then remove (or downgrade) your previous adfmin account. 80.40.144.68 (talk) 14:02, 9 August 2010 (UTC)[reply]

Absolutely nothing happens when yuo press Ctrl+Alt+Del twice at the logon screen :( Perhaps of more use is a way to edit the Windows registry through a version of Linux on the computer? 110.175.208.144 (talk) 09:45, 10 August 2010 (UTC)[reply]

Windows 7 installs by default include a recovery partition, you may be able to use this to add another administrator or unhide the existing one Nil Einne (talk) 15:50, 12 August 2010 (UTC)[reply]

What are some alternatives to Google Voice in the UK?

Preferably free... —Preceding unsigned comment added by 188.220.46.47 (talk) 16:19, 9 August 2010 (UTC)[reply]

internet

Resolved

If a computer is behind a router, you need to enable "port forwarding" in order for other computers to connect to it via the internet, say for example remote desktop or running a home server. But I don't understand then how a computer behind a router without port forwarding enabled can browse the web. Does google and wikipedia servers not send data to the computer browsing them? 82.43.88.151 (talk) 16:38, 9 August 2010 (UTC)[reply]

The computer has a local IP address. The router wraps the request so that the request will come back to it, and then unwraps it to see which local IP address originated the request. An example: You work in an office building that has the address 123 Some St. Your office number is 42. You list the address as 123 Some St, Office 42. The post office doesn't care about the office number, they just deliver to the building. Internally, your request is sent to your office. Web requests are the same. The remote sites send stuff to your router's address. The router handles the rest of the delivery. -- kainaw ™ 16:50, 9 August 2010 (UTC)[reply]

If you visit Wikipedia and you're behind a router, you download web pages from one of Wikipedia's IP addresses on port 80 (e.g., 208.80.152.2:80). The router then assigns your computer a unique port along with the router's IP address, like this: 54.84.93.20:4895. The source port number can be anything between 1024 and 65535.

You request a web page

Your computer		Your router		Wikipedia
192.168.0.3	→	54.84.93.20, port 4895	→	208.80.152.2, port 80

In other words, the router only has one public IP address, so it uses port numbers to differentiate the different local computers it is serving. This type of NAT is commonly called PAT (port-address translation). Wikipedia then has a return address (54.84.93.20:4895) and it can send replies to that address.

The only time you need port-forwarding is when some computer wants to request something from the router out of the blue. In that case, your router cannot make up a port number because it doesn't know what computer it wants it from. So, you have to set up forwarding to tell the router to route such requests to it.--Best Dog Ever (talk) 17:02, 9 August 2010 (UTC)[reply]

Thanks :) 82.43.88.151 (talk) 17:57, 9 August 2010 (UTC)[reply]

problems with DTS/WAV audio

I have a 5.1 receiver attached to my computer. The receiver recognizes all 5.1 audio embedded in "video" containers: DVD, DVD-Audio, MKV, etc. However, I've had no luck with "DTS CDs" / WAV files with DTS embedded. I've reviewed plenty of instructions out there on the web to make it work in: foobar2000, Media Player Classic (ffdshow), WinAmp. Depending on my configuration tweaks, I get stereo, white noise, or the receiver goes into DTS mode (DTS 96/24 actually) but there is no sound. (Configuration tweaks include resampling to 48khz; SPDIF used/not used; program-specific extensions.) Assuming I can't play these CDs/files "as is", does anyone know of a way to convert the audio to another format (extract channels and put them into a different container)? What could possibly be so special about 5.1 audio files that they just won't work, while the system handles 5.1 audio in video container formats just fine? I find the quality of DVD-Audio exceptional, and it works on my system; I assume I would get a similar experience from "DTS in WAV". Any advice appreciated. Riggr Mortis (talk) 18:19, 9 August 2010 (UTC)[reply]

How is it possible?!

Dear Wikipedians:

My ISP, Bell in Canada, will probably be one of the last ISPs on this planet to adopt IPv6, yet when I opened my bittorrent client today, I suddenly saw, beyond my wildest belief, two IPv6 addresses in my peer list. How is this possible???!!!

Thanks,

174.88.240.39 (talk) 21:09, 9 August 2010 (UTC)[reply]

IPv6#Automatic_tunneling. Unilynx (talk) 22:01, 9 August 2010 (UTC)[reply]

Long live Canadian ISPs! </sarcasm> 99.224.10.2 (talk) 21:03, 11 August 2010 (UTC)[reply]

IRC

Can people in IRC channels see my ip address? Can channel ops see it? 82.43.88.151 (talk) 21:36, 9 August 2010 (UTC)[reply]

It depends completely on what network you're connecting to so if you let us know which one we can advise better. Although normally it would be visible to all, some networks partially mask it and some of them offer custom hosts/masks. However, it's a fair assumption to say that regardless of the network, the server operators can always see your IP address. Hope this is of some help! ZX81 ^talk 22:19, 9 August 2010 (UTC)[reply]

Rizon is the network. By "server operators" do you mean the channel ops, or the actual server administrators who run the network? 82.43.88.151 (talk) 22:32, 9 August 2010 (UTC)[reply]

I've just connected to their network and as part of the login stage they automatically set your mode to include +x which partially masks your IP/hostname. In your case your IP 82.43.88.151 resolves to cpc2-croy8-0-0-cust1942.croy.cable.virginmedia.com so EVERYONE on that network would see your hostname as something like Rizon-E4096E55.croy.cable.virginmedia.com. The server operators however (people who actually run the physical servers) would be able to see your true IP address and it's also likely that any of the other Rizon staff could too. ZX81 ^talk 00:56, 12 August 2010 (UTC)[reply]

Why Ctrl + H?

Simple question that Google hasn't really been able to help me with. Ctrl + F is generally the shortcut for opening the "Find" dialog in a text editor (Notepad, Notepad++, and Word, among others). This makes sense, since "Find" has an 'f' in it. However, Ctrl + H is generally the shortcut for opening the "Replace" dialog and this does not make sense; "Replace" doesn't have an 'h' in it. So why Ctrl + H? Irish Souffle (talk) 21:43, 9 August 2010 (UTC)[reply]

Well Ctrl + R is usually "reload", so they can't use that. So with the only key which relates to "replace" already in use, perhaps they just picked a key at random for it. 82.43.88.151 (talk) 21:48, 9 August 2010 (UTC)[reply]

Also, Ctrl+F, Ctrl+G, Ctrl+H all do similar things - "Find", "Go to" and "Find and Replace respectively, so it makes sense to have these keys next to each other. 110.175.208.144 (talk) 21:56, 9 August 2010 (UTC)[reply]

In a similar vein, whilst "copy" begins with "C", "cut" and "paste" do not contain "X" or "V". AJ Cham 01:32, 10 August 2010 (UTC)[reply]

I don't know if this was the original rationale, but X looks like a pair of scissors, and V looks like an insertion mark. I'm old enough to have used Windows 2, where the copy/cut/paste shortcuts were ctrl-ins/shift-delete/shift-ins: these still work in many (most?) Windows applications, and my fingers still remember them....AndrewWTaylor (talk) 08:15, 10 August 2010 (UTC)[reply]

Ctrl+H has meant replace for much longer than Ctrl+R has meant reload. -- BenRG (talk) 07:54, 10 August 2010 (UTC)[reply]

I've no reference for this, but I believe Microsoft Word pioneered ctrl-H for replace. Comet Tuttle (talk) 22:08, 9 August 2010 (UTC)[reply]

Easiest way to remove a disc from a dead Wii?

Resolved

– « Aaron Rotenberg « Talk « 23:57, 9 August 2010 (UTC)[reply]

I have a Wii which was thoroughly bricked via lightning strike. I'm not too concerned about the console itself - there's plenty of Wiis (Wii-i?) to go around. I'm more concerned about the disc it currently contains, namely a copy of the limited-edition North American Metroid Prime Trilogy. I think the console's out of warranty thanks to sheer time. Google suggests a number of unsavory solutions, ranging from pulling it out with a paperclip to moving the internal clasps with a pair of tweezers to disassembling the entire console, all of which are spectacularly likely to scratch up the disc. Anyone have a better idea? « Aaron Rotenberg « Talk « 22:43, 9 August 2010 (UTC)[reply]

Uh... never mind. I managed to resurrect the damn thing by leaving it unplugged for a while and then plugging it back in. I have no freaking clue why that worked, given that the console showed every indication of a fried motherboard right up until now. And then I took the disc out, and it turned out to be SSBB. I had the Prime Trilogy disc sitting upside-down on the box this whole time. *facepalm* « Aaron Rotenberg « Talk « 23:57, 9 August 2010 (UTC)[reply]

Active Desktop (HTML content) on Desktop

I had been using Windows XP for a long time and now just upgraded to Windows Vista. In XP, one of my favorite features was Active Desktop which allowed me to put any web content as background on my desktop. I understand that it is not available in Vista anymore. My question is, is there anyway to make it work on Vista? Any settings, tweaks, or neat little programs which will let me use an HTML page as my background? I used to make cool java applets and then put them up as my background (much better than boring old static pictures). Just to be specific, I am using Windows Vista Ultimate and I need something that will display Java as well as allow me to interact with it. Thanks! -Looking for Wisdom and Insight! (talk) 23:53, 9 August 2010 (UTC)[reply]

Right. It's been replaced by the sidebar. If you go to C:\Program Files\Windows Sidebar\Gadgets you can view their source code. They're essentially web pages with XML settings files. So, you can still place web pages on your desktop by porting them to the sidebar. Here is a tutorial on them. Another method would be this gadget. It places images of your web pages on the sidebar.--Best Dog Ever (talk) 01:25, 10 August 2010 (UTC)[reply]