Wikipedia:Reference desk/Archives/Computing/2007 October 6

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.

October 6

Running IE

How would one run IE from a limited user account to a administrator account, if you know the password to both that is...ie. what does one need to type in "run" by windows to get it run IE as admin? Thank 209.202.45.185 00:42, 6 October 2007 (UTC)

I am not sure but if you know which version of Windows you are using that could help.

If you are able to right-click from the limited account, try right-clicking and then clicking on "run as" ... the rest should be a piece of cake. --Kushal^{Click me!} write to me 15:04, 7 October 2007 (UTC)

Firefox 2 Spell check

When typing a comment in a text field (such as this), I accidentally added the word "unverifyable" to its dictionary (The correct spelling of the word is "unverifiable"). I was about to clicked on the right word in the right-click menu, but not being as mouse savvy as I though I was, I click on the option JUST BELOW IT, which was "Add to dictionary." My question is how do I remove it from the dictionary? And why did they not at least put a separator between the suggestions and the add option, like what Microsoft® Word does? — Kjammer  ⌂ 01:34, 6 October 2007 (UTC)

Hi, if you look in your profile folder, you'll find a text file called: persdict.dat which you can edit. You might need to edit it when Firefox is not running. You should probably use a text-editor which can handle Unix line-endings. Wordpad will work if you're in a pinch. --Kjoonlee 16:20, 6 October 2007 (UTC)

How to find your profile folder: Profile folder at MozillaZine Knowledge Base --Kjoonlee 17:46, 6 October 2007 (UTC)

Thank you, fixed the problem. I was looking in the Program Files section for something like this and could only find DLL files related to spell check, which do not store any data. Although you'd think there would be a viewable list in the options menu. — Kjammer  ⌂ 22:40, 6 October 2007 (UTC)

Richard Stallman and Wikipedia

Hi all - thanks for any help with this query....

Has Richard Stallman ever written something or said something quotable specifically about Wikipedia that reflects his opinion of it? (perhaps particularly related to this article: http://www.gnu.org/encyclopedia/free-encyclopedia.html ) Thanks for any references, links and tips! 207.151.227.116 02:38, 6 October 2007 (UTC)

Image Scanner Upkeep

Hello. What are the unobvious tasks in maintaining an image scanner? I know that I must regularly clean the glass cover but what else? What if my light tube that moves up and down when scanning burns out? Thanks in advance. --Mayfare 03:08, 6 October 2007 (UTC)

I've never heard of a light tube burning out; I'm sure it is possible but I've never seen it, and don't think it is very likely to happen unless you are seriously mishandling the scanner. As for other maintenance, there really isn't any, though if you transport the scanner you have to re-calibrate it occasionally as it can get a bit off of alignment when moved around (this is done via the scanner software). --24.147.86.187 04:11, 6 October 2007 (UTC)

As the fluorescent tube or LED illumination thing ages, the color of it may go off. If it does burn out, they generally are not replaceable. The color changes, though, can be handled by the calibration that the previous editor mentioned. --Mdwyer 01:02, 9 October 2007 (UTC)

Energy consumption of the Internet

How much energy does the Internet use up? Worldwide and/or in a densely internetted country like the Netherlands. Not counting the endcomputers, because they may have been on anyway, even if there was no-one on the Internet. DirkvdM 06:02, 6 October 2007 (UTC)

Everyone should read slashdot :] Not exactly what you're looking for, but go to the article and read through the documentation; there are dozens of sauces that you can use to figure it out for yourself --froth^t 06:56, 6 October 2007 (UTC)

Wow, 5.3 % worldwide and 9.4 % in the US. Since I assume that energy waste in other fields will be higher in the US, that might be even higher in other western countries, especially in the Netherlands. But two thirds of that is due to the computers and monitors at people's homes, sort of like I expected, and it's not entirely fair to count that. But still, for the Netherlands it will probably be something like 5% of energy consumption, which is more than I thought. DirkvdM 17:56, 6 October 2007 (UTC)

IMO it's not really all that surpising. For every watt they're using, they waste 2 or 3 watts trying to cool the thing.. and my laptop uses a LOT of power for cooling- to keep it from overheating I use a USB-powered fan rig to pipe heat away from the bottom (my thinkpad tries to put out as much heat as possible through air cooling, and any extra it pumps directly through the solid metal bottom, so it's perfect for this kind of setup) as well as the built in internal air cooling system that runs air super fast through a little corner of the laptop and uses powered radiators to dump heat into that little fast-airflow space. I get about twice the battery life when I turn off internal cooling and unplug the fan rig --froth^t 18:50, 6 October 2007 (UTC)

I really don't think they use more energy in cooling than actually for calculation. All my fans are just rated for a few watts, and I have 6 of them, which is way more than the average; that's just around how much power my 250gig hard drive uses. --antilived^{T | C | G} 04:50, 7 October 2007 (UTC)

E-Mail ID

I want the E-Mail ID OF Tiffany Taylor.Now dont dismiss it as a borderline trollish question I am really crazy about her.218.248.2.51 07:57, 6 October 2007 (UTC)Hedonister

Her website says it's dgi.business AT aol.com. Good luck! --Sean 13:49, 6 October 2007 (UTC)

I would be really surprised if she actually ever sees any of the mail that goes to that address. If you get a reply - it'll be picked from a menu of pre-written replies. Forget it. SteveBaker 17:59, 6 October 2007 (UTC)

I would not be surprised. If I got nearly as much attention, I would probably do that too. --Kushal^{Click me!} write to me 15:01, 7 October 2007 (UTC)

Streaming music with wiki

Is there a way to stream music with wiki? For example using windows media player to play music in a wiki article? —Preceding unsigned comment added by 85.75.79.220 (talk) 08:08, 6 October 2007 (UTC)

Yes, in fact we've already got it integrated into many articles, like Eruption_(song). [Mac Δαvιs] ❖ 02:13, 7 October 2007 (UTC)

Hardisk Failure

Hi friends, My PC's been suffering from Hardnisk failure.My configuration is (Pentium D 3GHZ,1GB RAM(transcend),Mercury TV tuner card on PCI slot,160 GB SATA Segate HDD,965 INtel Mainboard,Windows XP SP2) In C Drive I have installed Windows 2000 and in D drive I have installed Windows XP SP2.The thing happened was., that last week I found one partition was not formatted and it was about 12 GB space.I have already divided HDD into 8 partitions.Using disk management in XP,I assigned a new drive letter "Z" and clicked format,and next to that second, my system displayed Blue screen or stop screen error telling that Plug and play device is in faulty cond,could be due to bad\faulty driver.Before this incident,my HDD was just doing superb.I never got such problems...I disconnected all other devices and again booted using that faulty HDD, but to my surprise, it showed the operating system choice menu such as windows XP and windows 2000.Also I was able to detect my drive in BIOS.Once I boot xp, the HDD is loading which I confirmed by its sound and vibration but when the "windows XP loading screen" appears,my system again displayed stop or blue screen error stating the same problem.This was the permanent case.Later,I plugged in the HDD as plug and play into another PC after booting and login to windows XP desktop in my friends PC.The device manager detects the drive, but within seconds it displayed stop screen error with same quote.This is my critical case.I changed HDD cables,power cord,I reset the BIOS and set for default values,but nothing helped out...Booting sectors in my HDD seem to OK,but I thing some clusters might have been damaged...I have huge amount of precious data with many not backed up.I cant give to data recovery centre since that will void guarantee or if I send it to segate,they will return new product and I will lose all data..Please help me solve this problem by any means...Thanks a thousand in advance... —Preceding unsigned comment added by Balan rajan (talk • contribs) 08:26, 6 October 2007 (UTC)

I suggest using a LiveCD for data recovery, such as Linux System Rescue CD or Rescubuntu, to recover it to another hard drive. Do it fast, since your hard disk seems to be near failure. Splintercellguy 20:27, 6 October 2007 (UTC)

Thanks But my hardisk is not at all loading.Only boot sectors seem to be fine...When my disk is being read by windows, the system automatically restarts.... —Preceding unsigned comment added by 122.164.57.137 (talk) 08:34, 7 October 2007 (UTC)

You can try burning a copy of a LiveCD version of GNU/Linux (I see a fellow Wikipedian recommended Rescubuntu) in another computer (I bet you could figure that out yourself) and then using that computer to transfer the data from your near to failing disk to the on in the computer. Your computer should not restart just because you have a failing (or even failed) hard disk connected to it unless you are booting from the failing hard disk. --Kushal^{Click me!} write to me 14:59, 7 October 2007 (UTC)

RAIDed external USB harddrive...

My dad wants to buy an external USB harddrive to back up all his important things. He's really, really worried about it crashing and only having one copy, so I suggested that maybe he should get one that had a built in RAID system (RAID 1, that is), because I imagined some USB drives came with that built in. There are, right? Can someone recommend one? --Oskar 09:38, 6 October 2007 (UTC)

They're pricey. This one looks perfect but it's way too expensive for only 400GB in RAID 1. Maybe this one, but there's a bad review. I don't know, just buy a nice low-end enclosure that comes with all the RAID parts, and then buy hard drives separately --froth^t 17:44, 6 October 2007 (UTC)

I don't think RAID is a good idea. It's too easy to imagine scenarios where both drives are lost at once -- power surge, controller electronics failure, earthquake depending on where you live. RAID mirroring is good for servers that need to be online all the time; when a drive fails, the server can keep running while you swap in a new drive. I don't think it's a good fit for an external backup drive that's only occasionally going to be switched on. I'd buy two drives from different manufacturers and alternate between them for backups, never having both plugged in at the same time. How much important data does your dad have? USB pen drives are dirt cheap these days. You could buy a dozen 4GB pen drives for the price of that Buffalo RAID array. -- BenRG 22:05, 6 October 2007 (UTC)

For critical data, I certainly wouldn't store it on the same computer - or even in the same building. RAID will save you from hardware problems such as a head-crash on one of your drives - but it won't solve the problem of you accidentally deleting a file you intended to keep - or of a virus or other malware taking over your PC and corrupting all of the drives in the RAID. That's why you need to mirror the data on another computer. But if your house burns down - it's not even enough to keep the data on two machines in the same building - so off-site backup is vital. My important stuff from home is mirrored onto the computer that does my web hosting once a week - as well as being copied onto my file server computer down in my garage every day. SteveBaker 13:47, 7 October 2007 (UTC)

If you are paranoid, maybe you can burn the data every few days on DVDs and mail copies to two reliable people living in abother county (or still better, another state or country). But I don't know how it would help for extended periods of time. --Kushal^{Click me!} write to me 14:52, 7 October 2007 (UTC)

I am not a computer scientist, but from what I have read, USB hard disk drives are painfully slow and are suitable only as a back-up drive. --Kushal^{Click me!} write to me 14:53, 7 October 2007 (UTC)

They are slow - but not painfully (well - depends what you are doing) - I have a beat-up old laptop whose hard drive died. I use one of those teeny-tiny 20Gb "FIREFLY" USB drive to boot the machine into Linux - and it's perfectly usable for stuff like web surfing and email that doesn't demand a lot of disk I/O. It's fast enough to play movies using mplayer too. So whilst I wouldn't recommend using USB for high-demand tasks, for basic stuff it works plenty fast enough. SteveBaker 18:54, 9 October 2007 (UTC)

problem of faint photos on internet viewing.

Hi, I am Tejraj Singh from Palghar, India, the problem I am facing with my computer is that when I connect internet, pictures/photos are seen faint not clear or say broken (like trying to hide fact in photos ) .this problem is with all internet sites that I had visited. Please help me how to overcome this problem on my computer, is their any short of software or else. Thanking you, Tejraj Singh.220.224.101.32 —Preceding unsigned comment added by 220.224.101.32 (talk) 11:38, 6 October 2007 (UTC)

Hello Mr. Singh. Do photos/pictures that you have on your computer also appear faint? If so, it may be either a problem with your monitor, or your graphics card.--Mostargue 11:57, 6 October 2007 (UTC)

It might be something as simple as turning up the contrast on your monitor. This can easily be tested with something like the shades of grey on the right in greyscale, but for your purpose that would have to extend to pure white (where you would then have to see all shades). Don't we have that somewhere? DirkvdM 18:02, 6 October 2007 (UTC)

This is a stretch, but it is possible that your browser is going through a proxy that compresses pictures to minimize bandwidth charges. On thing to check would be to see if the images are still damaged when you use the SSL version of a web page. --Mdwyer 00:59, 9 October 2007 (UTC)

microcontroller neuron

Are there microcontrollers designed specifically to simulate or duplicate the learning function of a neuron or does such functionality have to be programmed into a standard microcontroller, and if so, is there an algorithm for such programming? Clem 18:05, 6 October 2007 (UTC)

No - microcontrollers are not designed to act as artificial neurons. You could possibly run a Neural Network simulator on one of them. The algorithms for this are widely published in books on the subject - but without understanding the workings of them, you'd find it very hard to use them - so you're going to need to get hold of a book on the subject anyway. What application did you have in mind? SteveBaker 20:45, 6 October 2007 (UTC)

The Artificial neuron article does not show how an artificial neuron learns but only how an artificial neuron functions after it has learned. Clem 05:21, 7 October 2007 (UTC)

You need more than an encyclopedia article to cover this broad topic in enough detail to actually go off and implement a neural network. (You keep talking about a single neuron - I hope you realise that we need to talk about a network of many artificial neurons - typically all simulated in a single computer.) As I said previously, you need to get a book on the subject and be prepared for a significant amount of study. Training a neural network is by far the most complex part of the problem and there are a lot of techniques and gotcha's that have to be worked through. Very roughly, you start with a randomised set of weights for the neurons in your network. Then you present a typical input to the network and observe the output it generates. Then you reinforce the weights of the neurons that produced the output you wanted and you reduce the weights on the ones that messed up - and back-propagate those weights up the network towards the inputs. (Back-propagation alone is the subject of three chapters in the book I have.) Do this a bazillion times with different training samples, gradually reducing the amount of change you apply to the weights and eventually the network learns. However, there is a lot more to it than that - it's a huge field that's not easily summarized in the few pages that is appropriate to an encyclopedia article. The book I originally learned this stuff from ("C++ Neural networks and Fuzzy Logic") is wildly outdated and I couldn't recommend it - but there are plenty of more modern works which include sample code and worked examples. SteveBaker 13:37, 7 October 2007 (UTC)

Your description of how an artificial neuron (and artificial neural network) learns provides the general answer I was seeking in an encyclopedia article. Can you add your description to the artificial neuron article with perhaps some Pseudocode that illustrates the algorithm as well? Since the artificial neuron is the basic component, structural element or computing unit of the network I need an explanation that is comprehensive enough to be able to progress to an understanding of how an artificial neural network learns regarding further complexities such as the technique you mentioned of back propagation. Thanks. Clem 14:32, 7 October 2007 (UTC)

Generally, the implementation of an artificial neural network is a big array full of the weightings of the inputs to each neuron plus another array which contains the current state of each one and a third array that lists the connection pattern. Generally the connection pattern consists of an input layer - and output layer and some number of 'hidden layers' of neurons between input and output - with the connection pattern generally connecting all of the neurons on one layer to each of the neurons on the next (although MANY other variations are possible - and even commonplace). The main loop of your application shoves the input data into the input neuron's state array - then iterates over all of the neurons taking each of their inputs current state, multiplying by the weight for that input and then summing them to make the new current state for this neuron. Hence the whole system is (typically) just a simple 1D floating point array for the current-state of each neuron, a 2D floating point array for the weighting of each input of each neuron and a 2D integer array containing the array index of the neurons that connects to this input of this neuron (of course this could instead be a 1D array of structures or classes or something - but performance is the driving factor here - neural networks can take a long time to execute and cache coherency can be a big problem with the largest ones). Since there tend to be a LOT of interconnections (just as in a real brain, the number of interconnections is vastly more than the number of neurons) - the inner loop of the code is mostly fetching state, multiplying by weightings and summing. This being just a few lines of code, and given that it occupies almost 100% of processing time in running the network, it can be really carefully hand-optimised. This is a useful property of neural networks. SteveBaker 23:46, 7 October 2007 (UTC)

In terms of a practical application then suppose each input were a pixel of 32 bit color of a 600 by 800 array. Do there exist such neural network arrays for instance that might be built into the electronics of a CCD array which can be trained to recognize a set of similar pictures and then ring an alarm when anything like its training comes into view of the camera? Clem 00:17, 8 October 2007 (UTC)

Yes - that kind of thing is certainly possible. You'd need 600x800x3 = 1,440,000 input neurons (for red, green and blue for each pixel), some number of intermediate layer neurons and just one output neuron (you just want some kind of 'confidence' number on the output 1==It's one of the special images, 0==No, it's not a special image, intermediate values indicating degrees of confidence). You'd train the network by alternating between showing it one of your special images and tweaking gains to get a '1.0' output - and showing it a set of non-special images and training to get a '0.0' output. Then you'd be able to stuff a 'live' picture from the camera into the input neurons and observe the value of the output neuron. However, 1.4 million input neurons (and probably a few million intermediate neurons) will require an immense amount of training and a TON of computer resources to do that. The camera itself would need a pretty hefty computer to actually run that network in realtime. Remember, you'll want to connect every one of the input neurons to every one of the intermediate layer neurons - if there are 1.44 million of each, then that's 1.44² interconnections - which is 2 trillion multiply/add cycles. On even a 3GHz PC, that's going to take a L-O-N-G time to execute! Consequently, you probably can't have the algorithm looking at ALL of the pixels - I'd probably split the image up into (say) 30x40 grid cells and feed each neuron with the average of the pixel colours in its grid cell - so you only need 3600 input neurons - and you'd only have maybe 10 million multiply-add cycles. The resulting neural net wouldn't be able to recognise tiny features in the images - but it would at least run at a reasonable speed. But this is a tricky subject - you need to learn the technology (I'm far from being an expert) and spend serious time on trying different possibilities - neural networks are as much art as science. I used a neural network to recognise buildings in satellite photos - which meant that I fed a 128x128 pixel area to a neural net - and took a 128x128 output that indicated the probability of each pixel being a building. It worked well eventually - but it took me months of tweaking the resolution and number of hidden layers in the network. SteveBaker 13:37, 9 October 2007 (UTC)

As SteveBaker points out, the dimensionality of 800x600 images is really high. Each example picture is defined by by 1.4 million numbers, making most machine learners (such as neural networks) useless. The key is dimensionality reduction, such as the proposed method of reducing to a grid of 30 x 40. Another way to go about this is to use feature extraction. You define a bunch of ways to turn the image into a single number like "how many pixels are green", "how strong is the contrast", "how strong are the edges" and so on. If you use, say, 50 of these feature extraction methods, you get 50 numbers, instead of 1.4 million. You can then use these as inputs for you neural network (or other machine learners, support vector machines work well with this method). Another method of dimensionality reduction principal components analysis. Here, you look at a single picture as a vector of 1.4 million numbers. The algorithm then calculates (looking at the whole set of examples) another vector of 1.4 numbers where the first number is the most important, and so on. you reduce the dimensionality of the vector by just using the first n numbers and ignoring the rest. If you're going to do machine learning, you should also beware of overfitting. A good book on the principles of machine learning is 'Machine Learning' by Tom Mitchel.

risk 23:33, 12 October 2007 (UTC)

Bluetooth Webcam

Where can you get a bluetooth webcam?martianlostinspace ^{email me} 19:19, 6 October 2007 (UTC)

I don't think there are any. --froth^t 19:22, 6 October 2007 (UTC)

Really?martianlostinspace ^{email me} 19:26, 6 October 2007 (UTC)

Bluetooth isn't really made for high speed, high bandwidth transfers, 2.0 transfers only about 3 Mbit/s which isn't all that much. It's designed to transfer sounds and and calendar entries and stuff. I think that it would be pushing it to use it for a webcam. --Oskar 20:06, 6 October 2007 (UTC)

You might be able to just get a USB webcam and bridge it over a pair of Wireless USB adapters. Belkin apparently makes them but I can't find them. --froth^t 21:06, 6 October 2007 (UTC)

But even if you could - 3Mbit/s isn't enough for video. You can't magically get bandwidth from nowhere - so it couldn't possible work. SteveBaker 13:19, 7 October 2007 (UTC)

3Mbit/sec is more than enough for video. Sure, not full screen 30fps video, but mobile phones record video in 3gpp which have data rates belows 3Mbit/sec. I would guess the real reason is power supply and range. You most likely need to plug it in to AC power. And if you have a wire to power it, you might as well use power over Ethernet which would provide tons of bandwidth *and* power. And even if it was Bluetooth, it would have to be located somewhat close to your computer and paired to a specific. A WiFi webcam would seem more practical as it could be available to any computer on your network. --24.249.108.133 16:56, 10 October 2007 (UTC)

OK then, are there any wifi ones? What I mean is, is there a straightforward camera that sits on top of your computer, like a wired cam, but without the wires. That is, wireless. Not the type that some companies are doing, security type ones. Plain old webcam, like you might use on MSN, for example.martianlostinspace ^{email me} 18:41, 12 October 2007 (UTC)

python class method: does a function/method have a way to know it's own name?

Suppose you have the following python code:

   class person(object):
       def __init__(self):
           self.fn = 'homer';
           self.ln = 'simpson';
       def say_anything(self):
           try:
               this_func   = '__what_do_i_put_here__??';
               name        = this_func.__name__;
           except:
               print       "there was an error!";
               name        = 'say_anything';
           print self.fn +" "+ self.ln +" didn't "+ name;
   test    = person();
   test.say_anything();

The code produces:

   there was an error!
   homer simpson didn't say_anything

The goal is to produce the same output, but without the "error" ... which means I have to figure out a way to get a function/method to know its own name ... which means I have to figure out what to put in place of '__what_do_i_put_here__??'. Any suggestions? NoClutter 22:32, 6 October 2007 (UTC)

• See here. --Sean 14:46, 7 October 2007 (UTC)

Removing lyrics from songs

Using existing technology and software, is it possible to remove the lyrics from an MP3 song, leaving behind only the melody, so that one can sing the song themselves with only the melody playing? Thanks. Acceptable 22:59, 6 October 2007 (UTC)

Welcome to Wikipedia. You can easily look up this topic yourself. Please see Wikipedia:Reference_desk/Archives/Computing/2007_September_26#Karioke_Track. For future questions, try using the search box at the top left of the screen. It's much quicker, and you will probably find a clearer answer. If you still don't understand, add a further question below by clicking the "edit" button to the right of your question title. . dr.ef.tymac 00:41, 7 October 2007 (UTC)

As for a more nonautomated and quicker answer: No. [Mac Δαvιs] ❖ 02:07, 7 October 2007 (UTC)

Quicker? Perhaps. Correct? Doesn't seem so -- especially since the previous answer to this question includes side-by-side demonstration audio files that you can hear with your own ears ... if you follow the "automated" link already provided. dr.ef.tymac 05:41, 7 October 2007 (UTC)

Do I need to do this again? I used GoldWave again, on Rock Lobster (song) by The B-52's.. chosen because since the instrumental track is so distinct, goldwave separates it perfectly.

Convinced yet? It doesn't work with many songs, but if you have a good high-quality version (preferably lossless, though this is rarely available -_-) it turns out OK. Most of the time you have a ghost of the vocals left, but it's much quieter than before and it's perfectly easy to sing over. --froth^t 06:39, 7 October 2007 (UTC)

My son has been using the OpenSourced 'Audacity' audio processing program which has a pre-programmed filter for removing vocals. (He removes them from rock music so he can hear the guitar part better when he's learning to play a song). I presume it's using a similar technique to the 'invert-left-channel-and-subtract-from-right' - but it's got a bunch of adjustable widgets that let you play around to get the best possible vocal elimination. It's obviously not 100% perfect (nothing could be) - but it's amazingly good - you can cut the vocals down to such a low level that you can only just barely hear them during the quieter parts of the music and you can't hear them at all when the music is louder. SteveBaker 23:33, 7 October 2007 (UTC)

The leftover vocals are almost always the results of an echo effect applied after recording. In songs without the echo applied, the elimination will be perfect. Of course, songs without echoed main vocals seem to have lots of annoying backup singers. :P HYENASTE 22:03, 8 October 2007 (UTC)

Wow! That's really interesting. It makes perfect sense though. Thanks! SteveBaker 13:22, 9 October 2007 (UTC)

PSP Wifi

I have established a wifi connection on my PSP using my home computer's wireless access. I have noticed that the two devices use the same IP address. So...

QUESTION 1- If I were to, let's say, get banned from a site using my PSP (this has not happened, just an example), would I be banned from that site on my computer as well?

QUESTION 2- Let's pretend I have the Internet Service Provider SBC Global. Both of my internet connections, then, are provided by SBC Global. So, let's say that, on my home computer, I establish an SBC Global email account. Of course, I wouldn't have a pre-established SBC Global e-mail account on my PSP since I am simply using the connection. So, let's pretend that I set up a Gmail e-mail account for myself, that I will use when using my PSP. So, if I subscriped to something or ordered something using my gmail account, would anybody using the computer or SBC Global account ever be informed thru the SBC account of this? So pretty much, would the computer user be able to track the IP address and find not only the history of the computer, but also what has happened on the PSP?

QUESTION 3- Are there any extra fees thrown onto a wireless internet connection if there is a 2nd party using it via a PSP? —Preceding unsigned comment added by 75.23.84.129 (talk) 23:27, 6 October 2007 (UTC)

I'll assume you're in a residential setting, and your PSP was connected to a home broadband router with wireless capabilities. For the first, The IP address associated with your connection would have been banned, so any machines that use the router will be blocked too. For the second, no. For the third, no, it's your wireless router, isn't it? How can the ISP track how many machines are connected to the router anyway? Splintercellguy 01:19, 7 October 2007 (UTC)

is the psp or iphone a better media player (widescreen movies).

it seems the iphone in widescreen mode playing a movie is similar to a psp doing the same thing. which is better? (the actual experience though maybe its because of brightness, physical size, resolution, dpi etc) —Preceding unsigned comment added by 84.0.127.58 (talk) 23:37, 6 October 2007 (UTC)

I have no idea about PSPs. --Kushal^{Click me!} write to me 02:41, 7 October 2007 (UTC)

The PSP appears to have a wider screen, but I have never even saw an iPhone in real life, so I would say the PSP. —Preceding unsigned comment added by Jonathan (talk • contribs) 03:45, 7 October 2007 (UTC)

I know the iPhone sucks but it might be the best thing for amatuers like me. --Kushal^{Click me!} write to me 14:45, 7 October 2007 (UTC)

Keep in mind that you must prepare almost all media you want to view on your PSP (outside of UMD movies and DRM-free MPEG-4 -- like video podcasts and movie trailers). You must compress any video yourself. The iPhone has the iTunes store, which carries lots of movies, TV shows and video podcasts. So the iPhone is far more convenient if you don't mind paying a couple bucks for content. And personally, the ability to have the internet available no matter where you go (even if it is slow AT&T Edge service) is something that will change your life. --24.249.108.133 22:49, 7 October 2007 (UTC)

Surfing web without web browsers

Is it possible on a Windows XP/Vista to surf the internet without a web browser on your computer and instead using, for example, CMD? Thanks. Acceptable 23:43, 6 October 2007 (UTC)

Not sure about windows versions for these tools, but can you explain a bit more about what you need. There are some text-based web browsers, like the Lynx web browser. There are simpler tools for downloading, like wget, which are also text-based. For more complex stuff, there are scripts like python which could be used. --h2g2bob (talk) 00:07, 7 October 2007 (UTC)

See also cURL and links. --71.175.68.224 01:57, 7 October 2007 (UTC)

I think if a software allows you to browse the web, it is a web browser. --Kushal^{Click me!} write to me 02:40, 7 October 2007 (UTC)

You could hand-write raw ethernet frames in a hex editor.. that's about the only way you're going to be able to do it without using a networking program --froth^t 05:57, 7 October 2007 (UTC)

He didn't ask how to do it without a networking program, he asked how to do it without a web browser. It's very easy, as said below, just telnet into port 80 and write the HTTP request by hand. You don't even have to know what ethernet frames are, or that there even is any layer below TCP. JIP | Talk 05:41, 8 October 2007 (UTC)

OTOH, it has been agreed that you can surf the internet without web trousers (or even naked) 8-)[[1]] —Preceding unsigned comment added by 88.109.198.32 (talk) 08:27, 7 October 2007 (UTC)

LOL --Kushal^{Click me!} write to me 14:44, 7 October 2007 (UTC)

yes, you can telnet to port 80 and parse the site yourself. (imagine what it looks like). for simple sites you're familiar with, this isn't even insane. —Preceding unsigned comment added by 84.0.158.245 (talk) 20:58, 7 October 2007 (UTC)

Been there. Done that. Try this:

• telnet www.google.com
• when it connects (it won't say anything) type the next line
• get / http/0.9
• press enter twice
• You'll get the HTTP header (200 OK + headers), a blank line, and then the raw HTML.

Fun for the whole family! --Mdwyer 00:56, 9 October 2007 (UTC)