Wikipedia:Reference desk/Archives/Computing/2007 January 25

From Wikipedia, the free encyclopedia
Computing desk
< January 24 << Dec | January | Feb >> January 26 >
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.


January 25[edit]

seeking a service to provide low-/no-cost occasional dial-up access from Germany[edit]

Hello, I am looking for a recommendation by an experienced user, for a site that might offer free dialup service for German users, the way "Free UK" does, in England. Access to dial-up through local numbers is a service which can be extremely useful for people traveling. It doesn't have to be strictly free, but just a low- or no-cost occasional-use service for access to the web, via a phone line. Can anyone assist me? Thank you.

205.155.65.226 21:56, 24 January 2007 (UTC)[reply]

browsing history[edit]

Is there a way to delete the browsing history on wikipedia from my computer? —The preceding unsigned comment was added by 24.15.67.23 (talk) 00:58, 25 January 2007 (UTC).[reply]

If you're using Internet Explorer 6, go to Tools->Internet Options. Click the General tab, and click the Delete Files button and Delete Cookies button. If no one tackles Firefox by the time I get off work, I'll explain that one too. -- Kesh 01:24, 25 January 2007 (UTC)[reply]
In Firefox, Tools > Clear Private Data > tick all that apply (probably just Browsing History, Cache and Cookies) > Clear Private Data Now. -- Matt Eason (Talk &#149; Contribs) 01:28, 25 January 2007 (UTC)[reply]
In Safari, pull down the History menu and choose "Clear History". Dismas|(talk) 06:40, 25 January 2007 (UTC)[reply]
If u just want to delete some of the history, for firefox, Hold ctrl+h and delete the ones u don't want. do same for IE --|K.Z|Z.K| Do not vandalize... 07:01, 25 January 2007 (UTC)[reply]
If you connect through a proxy server - for example if your internet connection is throught your school or university - then they may keep a record of your activity. Also, the website you connect to will log who connected to it by storing your IP address. There's no way to delete either of them. --h2g2bob 15:52, 25 January 2007 (UTC)[reply]

Linux installation.[edit]

Is it possible to install a Linux distribution on my computer then download the wine emulation program and still run all of my old windows programs without backing them up.

Also if I install Linux will it format my harddrive and how do I save Windows Xp to a disk so I can reinstall it just incase.

Oh and I have never used Linux before, what are the system requirments and would I need to download drivers. —The preceding unsigned comment was added by 67.67.217.21 (talk) 01:05, 25 January 2007 (UTC).[reply]

1) Not necessarily all of them. See WINE.
2) If you don't have a Windows installation disc, you don't. Don't tell me you pirated a copy! ;) That said, most modern Linux installers will create a seperate partition, so you can keep Windows and Linux both on the drive.
3) It varies from distribution to distribution. See List of Linux distributions to see how many you have to choose from. Some can run on old 386 boxes, others need a modern machine to run well. It depends on your needs. -- Kesh 01:26, 25 January 2007 (UTC)[reply]
1) In most cases you will find better alternatives that run natively. But yeah, wine doesn't run every single programme for Windows.
2) Dual booting is simple and most dists will do it on installation.
3) Generally, no. If everything doesn't work out of the box it is considerably harder to try and fix it. The only driver I need to install on my Ubuntu is the NVIDIA graphics card driver to enable 3D acceleration, and I think they are going to include that and the Radeon driver in the next release, Feisty, so basically no seperate driver needed in the near future, as most things come built in. --antilivedT | C | G 04:55, 25 January 2007 (UTC)[reply]
1. WINE doesn't work very well at all, but there are free alternatives to most things. Use a LiveCD (which doesn't touch your hard drive) and mess around with the programs on it. Most of those programs can also be used on Windows.
2. If you repartition your hard drive, it will wipe everything from it, so I'd suggest buying a cheap new hard drive and install Linux on that without touching your Windows drive.
3. Minimum requirements depends on what you're doing. There are some distributions which allow for really low specs, but for the main distros it'll work fine on a computer a few years old.
--h2g2bob 15:48, 25 January 2007 (UTC)[reply]
1. Actually, WINE has worked fine for me for nearly every program I've used with it, (GrabIt, QuickPar, and a few others), check here at the AppDB to see if what you'd need has an entry.
2. You can repartition and still keep your data with the free GPartEd, as well as other tools, although it is still easier to give Linux its own HDD, that may not be an option. Be aware, though I and many others have used GPartEd many times with no issue, I would still backup anything really important to you, as theres always the small chance of something going wrong, or a brownout, etc.
Cyraan 17:52, 25 January 2007 (UTC)[reply]
Actually, gparted is not the safest way to repartition. I suggest using ntfsresize from ntfsprogs.--68.250.41.99 00:08, 26 January 2007 (UTC)[reply]
gparted uses ntfsresize. And I'd say that it would be extremely dangerous to sit a newbie down with ntfsresize and let him mess with the partition tables. --frothT 04:37, 27 January 2007 (UTC)[reply]
gparted is a script which may or may not work. From experience, I've had it adjust the superblock on /dev/sda<partnumber> and then try to write the partition table to /dev/hda - which broke things rather nicely.--Frenchman113 on wheels! 16:15, 28 January 2007 (UTC)[reply]

I have a blogging question.[edit]

After I post my blog, it takes me to a screen that shows all previous posts and how many comments I have received, etc. It also has a column called "AggView." I was wondering if anyone knew what that ment. Thanks! --Zach 04:01, 25 January 2007 (UTC)[reply]

Aggregate view! It probably aggregates your entries up, possibly with comments. X [Mac Davis] (How's my driving?) 04:31, 25 January 2007 (UTC)[reply]
I am not too sure about that, One day it can say 20, the next it says 0. Thanks for your help! Any other ideas? --Zach 03:15, 26 January 2007 (UTC)[reply]
What blog are you using? -- Kesh 03:46, 26 January 2007 (UTC)[reply]
Its actually a website SteepleMedia.com (Community Server 2.0) --Zach 11:23, 26 January 2007 (UTC)[reply]

MAC Question[edit]

I have MAC OS X v10.4 (Tiger) and was wondering how to get v10.4.3

Is it available on a free download, as all the shops round here say they don't sell it.

Sorry, disregard the question, I just found out I have v10.4.6

CCLemon-ここは寒いぜ! 09:43, 25 January 2007 (UTC)[reply]

Genearlly, all the sub-sub version numbers (The "Y" in 10.X.Z) denote the Apple version of Windows Service Packs and can be downloaded from Apple's website or obtained via the "Software Update" program. 68.39.174.238 21:43, 26 January 2007 (UTC)[reply]
Speaking of Software Update, I suggest you run it. The latest version is 10.4.8 (http://docs.info.apple.com/article.html?artnum=304200), and 10.4.9 may be out soon (http://www.macrumors.com/2007/01/17/apple-seeds-another-mac-os-x-10-4-9-8p2120/). 68.39.174.238 21:46, 26 January 2007 (UTC)[reply]

Origins of Mobile Banking[edit]

I am currently doing some research and am struggling to find out which bank and in which year launched the first mobile banking solution. I am specifically looking to find out who launched the first sms banking solution but would also be interested in the first wap solution. Any information would be greatly appreciated. Thanks Y. —The preceding unsigned comment was added by 195.137.116.229 (talk) 10:11, 25 January 2007 (UTC).[reply]

Which country? Worldwide? And you might have to define "sms banking solution" a bit better. Is checking balances via SMS sufficient to qualify? Or transferring funds between your own accounts? Or transferring funds to someone else's accounts?--inksT 10:31, 25 January 2007 (UTC)[reply]

Worldwide - checking balances and by using SMS should be fine. Thanks Y.

GPG passphrase cracker.[edit]

I found one of my old GPG keys, and I'd like to use it. Alas, I seem to have forgotten my passphrase. I have a quick-and-dirty Perl script executing gpg a bazillion times with each possible passphrase, but it's pretty inefficient, and won't finish in a reasonable amount of time. Is there a key-cracking utility somewhere out there which has a faster inner loop than that? grendel|khan 15:11, 25 January 2007 (UTC)[reply]

The entire point of encryption is to keep people from cracking the codes. Especially if you used a fairly strong encryption scheme. I'm afraid you're stuck either brute-forcing for a long time, or seeing if perhaps you stored your passphrase somewhere else. -- Kesh 23:50, 25 January 2007 (UTC)[reply]
You can speed up the search by optimizing it for your typical length, characters, etc. Most brute force scripts start at 1 character long passwords and work their way up, using all the special characters and numbers as well as capitalization variation to try every possible combination. This is inefficient since most people are not that wary of weak passwords, so you might try some simple optimizations to reduce the crack time. By the way, the technology behind even old versions of GPG is still considered to be very secure, so you're stuck with brute forcing it. Droud 00:23, 26 January 2007 (UTC)[reply]
Try john, it's probably more optimized than your "quick 'n' dirty" script --frothT 18:27, 26 January 2007 (UTC)[reply]
Also, do you remember ANY part of your phrase? Even how long it was or one character that was included in it? Anything could be used to optimize the search. 68.39.174.238 21:47, 26 January 2007 (UTC)[reply]

Thanks, everyone. I remembered the basic layout of the password, and iterated over a large set of variations on that, which eventually extracted it. Close call, though. Phew! grendel|khan 00:41, 27 January 2007 (UTC)[reply]

Preloading an image for a div background in javascript[edit]

Hello. I'm making a page where a mouseover changes the background of a DIV using a simple Javascript function. Simple stuff, except I can't figure out how to get it to preload the image before it tries to replace it. As it currently stands it will change the image, but first the DIV background goes blank for a second while the browser downloads the new one. I'd rather it just keep with the old one until it has fully loaded the new one, and then switch. Here is my current code, which I thought would accomplish this, but apparently not. Suggestions?

function changeimage(id,imagepath) {
	newimage = new Image();
	newimage.src = imagepath;
	document.getElementById(id).style.backgroundImage='url('+imagepath+')';
}

I thought that by loading it in the Image first it would pre-load it, but it doesn't seem to work. Perhaps I need to assign it to the element differently? I'm not sure. Any tips would be appreciated. (Actually, now that I check, the elements in question are not DIVs, strictly speaking, but LI, but that shouldn't make a difference?) --24.147.86.187 15:57, 25 January 2007 (UTC)[reply]

NOTE: I tried putting the preloading in at the top of the script (outside of the function) and it works OK when it is there. But I would like to be able to selectively pre-load, ideally, so I wouldn't have to preload all rollover images before use. If that makes sense. --24.147.86.187 16:05, 25 January 2007 (UTC)[reply]
The image will be loaded when you make the new Image with the src attribute. Some browsers will cache an image while others will discard it as soon as you leave the function it was created in, because it is a local variable. As with all javascript stuff, make sure there's a noscript tag about for people who don't have javascript enabled (like me!). Also think of implementing rollover effects using css rather than javascript. --h2g2bob 18:20, 25 January 2007 (UTC)[reply]
I think what I'll end up doing is preloading in Javascript (which works ok) but changing the image itself with a :hover attribute. --24.147.86.187 00:38, 26 January 2007 (UTC)[reply]
Actually, what I ended up doing is loading the preloading script after the page has displayed (by using a setTimeOut with half a second or so as the interval), which works quite nice. The page loads up very quickly, but immediately after doing so it begins to precache some of the rollover images. By the time the prospective visitor starts to roll over something it should have already loaded. --24.147.86.187 00:51, 26 January 2007 (UTC)[reply]
I think what is happening is that the browser is preloading the image at the same time you are changing the background image. I believe the browser does not wait when you set newimage.src (if it waited, it would cause an obvious pause when loading the page when you preloaded at the top of the script); thus, when it gets to the next line (setting the background image), the image hasn't been downloaded yet. --cesarb 20:42, 25 January 2007 (UTC)[reply]
I think this is partially correct. When you preload the image at the top of the page it definitely waits until it is preloaded (it takes a second longer for the page to even start to render). However when I call it in the function itself it seems to try and start loading before it assigns it -- not quite quick enough. I'm not sure there's an easy way around it. I could probably simulate the effect I want by actually having two DIVs, one under the other, and just never getting rid of the original (bottom) one, but I don't know if that will look right... --24.147.86.187 00:38, 26 January 2007 (UTC)[reply]

If it's not an actual tiled image (or is only so in one dimension) you might consider using a single image with both pictures in it and setting the "background-position" attr to shift the different parts into view.

Isolating a segment of a LAN[edit]

Is it possible to use a second router to isolate a part of a LAN (TCP/IP)? I'm attempting to set up a combination wired/wireless network such that wireless connections to not have access to the wired connections, but the whole network has access to the internet. Right now the entire network is through a broadband router. I have a second router available. Can I do what I'm attempting to do? If so, how? Thanks in advance for any help. –RHolton– 17:03, 25 January 2007 (UTC)[reply]

Dunno about using a 2nd hardware router, but if you've a PC with 3 network interfaces, you could use Linux with Iptables to route between eth0 and eth2 & between eth1 and eth2, but not between eth0 and eth1 (where eth2 is the WAN/broadband router and eth0/eth1 are the wired & wireless networks). You could probably do the same on other OSes too. Davidprior 19:52, 25 January 2007 (UTC)[reply]
If you're using a WRT54G, you can turn on "AP Isolation" under Wireless--Advanced Wireless Settings, which prevents wireless clients from talking to each other, and may or may not prevent them from talking to wired clients. Also, if you use the second router for the wired clients, and manually configure it to use the wireless router as its primary gateway, you'll be able to connect to the internet or whatever from the wired clients, but you can leave out routing information for the wireless router, so that you've effectively hidden the wired network behind a NAT layer. grendel|khan 21:34, 25 January 2007 (UTC)[reply]
If you have two routers (one wired one wired/wireless) and one gateway (your cable modem) and no other fancy way of achieving it, then simply plug the wireless router into the cable modem, the wired router into one of the wired ports on the wireless, and turn them all on. The wired router will protect its clients from any incoming traffic, but they will both still share your internet access. --66.195.232.121 22:32, 26 January 2007 (UTC)[reply]
My homenetwork is isolated in 3 parts. I'm a cheap bastard :) so i did this by buying a old pentium and 3 ethernetcards for almost free. If you install openBSD or a security-targeted linux distribution on it you can make a very secure firewall/router from it (or you can just use your 2nd router if you already have it (don't expect much security from routers made for home-users)). When you want multiple parts in your LAN, you will want to use subnet's. Just see that they don't overlap. I'm guessing you don't need more then 254 ip's in 1 subnet so it's easiest if you number your subnets like this: 192.168.0.0/24, 192.168.1.0/24, 192.168.2.0/24, ... /24 is the netmask, this is the same as 255.255.255.0. I don't visit this page often, so post something on my userpage if you need more info Garo 10:12, 27 January 2007 (UTC)[reply]
Thanks, all, for the feedback. I ended up plugging the wireless router into the cable modem, set up as a DHCP server for addresses 192.168.1.*. I then plugged the other (non-wireless) router into the first router, with a fixed address of 192.167.1.10 and as a DHCP server for 10.0.0.*. I created a static route on the first router for the second router. Everything seems to work, though I didn't realize that NAT would cascade that way. From just basic tests, it appears that I cannot access nodes on the wired router from a node on the wireless--which is exactly what I was hoping for. Anyone know what sort of security holes might be present? -–RHolton– 20:26, 27 January 2007 (UTC)[reply]

wikipedia or social networking research subjects[edit]

I'm trying to think of some worthy topics to research under those areas. This will be quite the project. Can anyone suggest ideas? Thanks.

132.239.90.231 19:54, 25 January 2007 (UTC)[reply]


An Econ prof. of mine is doing his PHD on online markets (how efficient they are, and ways of predicting pricing trends etc.), but I can't find anything he's published online. You may also want to check out wwww.wikinomics.com. The site doesn't have much content, but the book looks interesting. It also looks like it could be just some guy trying to milk a social trend w/o any real insight.
I tend to be a "Markets" guy so I find it interesting to see WHEN different ones were acquired, and for how much (and $ per member, $ per member-hour spent online per month etc.). You can predict the peak of their popularity when that last figure stops increasing.

NByz 20:33, 25 January 2007 (UTC)[reply]

This for sociology? Any subject? Inherent differences between wiki and meatspace communication and how they underlie all/only parts that meet [Find out and describe a certain rule] rule would be useful. Suggest asking this on Humanities. 68.39.174.238 21:54, 26 January 2007 (UTC)[reply]

Dell Inspiron 5100 "Tap to Click" on the Touchpad[edit]

I've been struggling with this on and off for a while. My Dell Inspiron 5100 Laptop(Running windows XP Home) keeps enabling the "tap to click" function for the mouse touchpad on its' own. Sometimes within 10-20 minutes of disabling it in the mouse section of the control panel.

The problem tends to happen in bunches (today, for example, it has happened 3 times, and a couple yesterday, but not at all in the prior week or two. It has happened in bunches every couple of months since I've had this laptop (maybe 2-3 years)

Anyone had similar problems on similar Dell laptops?

NByz 21:09, 25 January 2007 (UTC)[reply]

Have you gone to Dell's website and downloaded the latest Synaptics touchpad software? [1]. If that doesn't work, Synaptics's site has a newer, generic version, though it's not guaranteed to work as well as the Dell-supplied one. [2]. Also, might there be some odd key combination that you're pressing to activate it? -- Consumed Crustacean (talk) 21:30, 25 January 2007 (UTC)[reply]