Wikipedia:Reference desk/Archives/Computing/2009 December 3

= December 3 =

How to prevent automatic re-directing on a website?
A website was created using a service, Moodle. However, the service provider was later switched to WetPaint. But, whenever the administrator tries to view the new website on their laptop they are unable to, and all that shows up is a generic 'demo page' for Moodle. The new website works fine on all other computers. Deleting cookies and setting firefox to alert you when re-directs take place haven't fixed this. The same thing happens with IE. Does anyone have any ideas on how to fix this? Thank you. —Preceding unsigned comment added by 60.234.239.245 (talk) 00:38, 3 December 2009 (UTC)


 * Do you mean "view the new website" that you are going to yourwebsite.com and it is going to the wrong place? If so, and you have set up the DNS correctly, then it probably means that whatever local DNS server your computer is connecting to is just not updated, and probably will in a day or two. Sometimes this kind of thing takes a day or so to percolate through the system. But how long has it been? --Mr.98 (talk) 03:14, 3 December 2009 (UTC)


 * On the laptop, go to Start --> Run... --> cmd --> ipconfig /flushdns. Then, try again. The DNS Helper Service caches DNS resolutions, which can become out of date. If that doesn't work, then try typing net stop "DNS Client".--Drknkn (talk) 03:37, 3 December 2009 (UTC)

Maximized embedded flash video minimizes when I click elsewhere
I have a dual-screen setup (Windows 7, GeForce 9600 GT with 2 BenQ 16:9 monitors if that matters...) and I like to use the right screen for video.

Unfortunately, when I go full-screen on many embedded video clips (like ones on CNBC.com - which uses Adobe Flash Player 10 in Google Chrome)on one monitor, then click anywhere on the other monitor, the video goes back to it's original size.

It would be nice if I could watch a video like that full-screen on one monitor, while surfing the internet on the other. Any ideas?NByz (talk) 02:24, 3 December 2009 (UTC)


 * I don't have a solution, but can offer a work-around. I had a similar problem using full-screen mode on one screen of a two-screen setup.  My solution was to simulate full-screen mode by dragging the corner of the window off the edge of the screen.  This is easiest if the screen layout is offset, like so:

+-+ |    |+-+ |     ||     | +-+|     |        +-+


 * I have CRT monitors, so I can also adjust the horizontal and vertical scaling and panning, so that I don't see the window edges. With most LCD monitors you lose this ability, unfortunately, so you might have to settle for leaving two adjacent sides of the window still showing. StuRat (talk) 06:11, 3 December 2009 (UTC)


 * Try this — Matt Eason (Talk &#149; Contribs) 11:54, 6 December 2009 (UTC)

How people look up information
I am trying to find statistics on how people look up information - by library, internet, buying books etc. In other words how people try to find information to answer something. I can't think of a way to phrase the search query or find to find any relevant statistics. I am writing a paper and need to substantiate the claim that people prefer to use Wikipedia, search engines, google, online databases etc as opposed to going to a library for their information retrieval needs. I know this is true.. but I need statistics to back it up. And don't know how to find them. Thanks for any help. —Preceding unsigned comment added by Baalhammon (talk • contribs) 03:26, 3 December 2009 (UTC)


 * You can search for prior research using Google Scholar. Surely somebody has collected statistics on exactly what you are working on.  Some helpful search terms might include "information retrieval" or "information systems".  I found Interaction with Texts: Information Retrieval as Information-Seeking Behavior, which has some helpful conceptual overviews and may lead you to exactly what you're looking for.  Nimur (talk) 04:31, 3 December 2009 (UTC)


 * I appreciate the heads up - but I already have a conceptual understanding of the topic, I need hard statistics. I wouldn't be posting here unless I already looked using those keywords - what comes up is exactly what you linked to, theoretical papers. I need statistics that demonstrate how many users use Wikipedia, search engines, online databases, other web 2.0 technologies versus library visits. I think its obvious to most people that this is true.. but without the data I can't make the claim in good conscience. —Preceding unsigned comment added by Baalhammon (talk • contribs) 05:53, 3 December 2009 (UTC)


 * The reason people don't go to the library is that it requires a trip. However, there are other sources of info at home that might offer more competition for the internet; like dictionaries, the Bible, medical texts, and even asking a family member.  Are you including these ways to "find stuff out" ? StuRat (talk) 06:02, 3 December 2009 (UTC)

Remastering an MP3
OK, so I have a Shirley Temple song on my playlist. Since it was a bit messed up (i.e. audio levels, static, etc.), up to how much extent can it be restored or remastered using software? Blake Gripling (talk) 06:43, 3 December 2009 (UTC)
 * There's a lot of audio editing software that can remove static and let you adjust levels. Audacity is free and may do the job, but for more demanding work there are many commercial packages too, such as Adobe Audition.  See also List of free software for audio, Category:Digital audio workstation software.  You'll have to experiment to see how well it can be fixed, as this will depend on the type of noise and how badly it is degraded. --Pleasantman (talk) 13:02, 3 December 2009 (UTC)


 * Note that this is a non-trivial task... it will probably not be easy unless there is a uniformly wrong thing (e.g. total song wrong volume, or wrong pitch). Otherwise it will take a lot of careful manipulation, and even then might not sound much improved (I find trying to improve audio often leads to it getting pretty swampy, even if it is something that should theoretically be easy, like removing a constant humming noise). --Mr.98 (talk) 22:27, 3 December 2009 (UTC)
 * Removing an unknown noise signal is a theoretically hard statistical signal processing problem. The trouble with these sorts of interfering sounds is that they come from real noise sources - so the sort of theoretical approximations like constant frequency hum, or gaussian distribution noise, or smoothly varying amplitude, are all out the window.  From the standpoint of looking at a waveform, the conventional assumption that it is a mere "superposition" of the desired music with an additive noise source becomes less useful in practice, (since the noise source has virtually no useful descriptive parameters).  Approximating the noise so it can be subtracted out requires many assumptions about the noise shape; alternatively, you can work on a few seconds of the audio at a time and tune your parameters manually.  As has been noted above, this kind of manual remastering will require a lot of time and effort tweaking parameters, even if you have very sophisticated tools.  The end result may still be noisy.  Conversely, sometime a simple equalizer or level adjustment is all you need to improve the quality - and those operations are fairly straightforward one-shot processes.  Nimur (talk) 15:12, 5 December 2009 (UTC)
 * It is pretty much so, although I often wonder on how Legend Films pulled off such a restoration job on Shirley's films audio-wise. They did a good job at the deep bass tones and stuff when I watched Rebecca of Sunnybrook Farm, but like what you said earlier, it won't be of trivial matter considering the age of the film and how degraded the footage was. Blake Gripling (talk) 02:26, 8 December 2009 (UTC)

return
In a function in C, one uses return value to return a value back to main. What happens when there is no value following return? Does it return nothing or all the values?

for example:

-- penubag  (talk) 09:10, 3 December 2009 (UTC)

As the declaration of the function says, it returns a void - i.e. nothing. If you were to try to use the return value:

then you would get a compile error. --Phil Holmes (talk) 09:39, 3 December 2009 (UTC)


 * If the return type of a function is void, it does not return a value, and the only valid return-statement is on the form return;. Otherwise, the return-statement must contain an expression convertible to the function's return type. In any case, the function main shall have a return type of int. If your program does not match these requirements, anything can happen. decltype (talk) 09:43, 3 December 2009 (UTC)
 * Ahh, I see. So it has the same effect as return void or not having anything?-- penubag  (talk) 10:08, 3 December 2009 (UTC)
 * Well yes, except that "return void;" is not valid. So for a function with void return type, "return;" simply ends execution of that function and returns to the calling function. decltype (talk) 10:29, 3 December 2009 (UTC)
 * It may be worth saying that you don't need to put a  at the end of a   function; if you simply hit the end of the function body (the last  ) you'll exit the function just fine.  You use   in a   function when you want to exit it other than at the end (which some, but not all, programmers think is a bad thing: see for example ).  (You can also return early from a function that returns a value; you must supply each   statement with a value, though they need not be the same.)  --Tardis (talk) 16:13, 3 December 2009 (UTC)
 * Yes, if you follow the Single-Entry Single Exit methodology, you are unlikely to have any return statements at all in your void functions. There could still be one at the very end, but I believe most people would consider it redundant. <span style="font-family:monospace, monospace;">decltype (talk) 15:02, 4 December 2009 (UTC)
 * From what I can remember,  is legal in pre-ANSI C, but not necessarily in ANSI C.   is legal in both,   is illegal in both. ("Illegal" meaning "causes undefined behaviour", meaning it might work all perfectly under some particular operating system, but there's nothing whatsoever guaranteeing it won't make the entire computer explode under another operating system.)   has three standard return values: ,   and  , of which the first two mean the same thing. Any other return value is implementation-dependent. Falling off   without returning anything counts as returning  , but   is the only function guaranteed to do that. For any other function returning anything other than  , falling off the function without returning anything causes undefined behaviour.  J I P  | Talk 20:56, 3 December 2009 (UTC)


 * What does "all the values" mean? Also, it's not at all the question you were asking, but it's probably important: your array   probably has length  .  Yet the first time through the loop what you try to print is , which is a bad thing.  --Tardis (talk) 16:13, 3 December 2009 (UTC)
 * Thanks for this! -- penubag  (talk) 20:36, 5 December 2009 (UTC)

Symbols, colors, fonts won't copy into email
I'm reposting this question from Vchimpanzee, who asked some follow-up questions after the thread was archived. I've added a question to VChimpanzee for more detailed info. --NorwegianBluetalk 15:55, 3 December 2009 (UTC)

Referring back to this question:

I now find myself unable to copy and paste this information into emails. Is this something new in Internet Explorer 8?

I do recall warnings that I was telling the computer to do something unsafe and that was stopped, but I hardly see where just the symbols would cause a problem. I do remember video or something else moving and I was told don't use Explorer. I stay away from Firefox and use plain text when possible, but copying the symbols manually is such a pain. Vchimpanzee ·  talk  ·  contributions  · 20:43, 24 November 2009 (UTC)


 * No. The problem is not IE8. Probably, your email client is set to using just plain text. If you switch to using html, it should work (at least sort-of-work). See E-mail. --NorwegianBluetalk 21:14, 25 November 2009 (UTC)


 * I'm not sure what that means. I just go to the web site of the email service and sign in (#Webmail in the above article). It doesn't seem to matter what computer or what address, but there is a problem with this now.


 * I'd rather do plain text but sometimes don't because copying the symbols is a pain. Vchimpanzee ·  talk  ·  contributions  · 21:28, 27 November 2009 (UTC)


 * Update: while I couldn't see the text with the symbols and fonts, it did go through. Vchimpanzee ·  talk  ·  contributions  · 20:47, 2 December 2009 (UTC)
 * To get a more specific answer, it would be helpful if you stated which email client you use, or, if you use a webmail service, which service you use (gmail, yahoo etc). It would also be helpful if you provided a link to a web page with contents that you want to copy and paste into your email. --NorwegianBluetalk 15:55, 3 December 2009 (UTC)

Fun Ti 83 Calculator programs
What are some cool TI 83 calculator programs?Accdude92 (talk to me!) (sign) 18:03, 3 December 2009 (UTC)
 * . --Michael WhiteT&middot;C 18:36, 3 December 2009 (UTC)
 * ticalc.org is the most comprehensive site to find programs. I remember playing a multiplayer Bomberman type game back in high school.  I think it required MirageOS which can be a little tricky to work with. Caltsar (talk) 18:39, 3 December 2009 (UTC)

Wonderful Special Effects on Wordpress
Just have a look at this wordpress blog. See the wonderful "snowfall". How can I do same on my own wordpress ?

Jon Ascton   (talk)  18:11, 3 December 2009 (UTC)
 * The first google hit for "" answers the question, but unless your blog is about falling snowflakes I'd file this under "blink tag". --Sean 18:59, 3 December 2009 (UTC)


 * Like this http://www.sjbaker.org ? Well, go to  - copy the javascript from that paste into your own web page right below the BODY tag at the top of the file.  Then find the place where it says "var snowsrc=" ...and delete the URL and change it to just "snow.gif".  Then make a small GIF image of a snowflake (or just a white circle - like I did) with a transparent background and place that into a file in the same directory as your web page...and you're done! (If you're feeling super-lazy, you can steal http://www.sjbaker.org/snow.gif - which I hereby release under the GFDL and Creative commons yadda yadda yadda (yes, it's a 16x16 pixel white circle, created with all of my very best, professional graphics skills so it's a valuable contribution to society...enjoy!) SteveBaker (talk) 19:14, 3 December 2009 (UTC)
 * (Revised: I made a spinning snowflake - it's even more annoying than the circles!) SteveBaker (talk) 20:17, 3 December 2009 (UTC)

Function templates with Microsoft cl.exe compiler
With the following lines of code:

For the line T max(T x, T y), I get the error c2027 stating "use of undefined type 'T'." I've tried putting "typename" in front of all possible permutations of the Ts on that line, but to no avail. What format does the Microsoft compiler require for function templates? 20.137.18.50 (talk) 19:52, 3 December 2009 (UTC)


 * I'm guessing a little here, but they are probably guarding against pathological cases. Declared in the manner you describe, the size of the return type (and the parameter type, which needs to be copied) could be near infinite, which could cause problems. If you declare it to return a T&amp;, it should work per this KB article, assuming I'm reading it right. &mdash;ShadowRanger (talk 21:33, 3 December 2009 (UTC)
 * That KB article is about code like <tt>template<class T> foo {...}</tt>, which is not legal C++, not <tt>template<class T> T foo {...}</tt>, which is legal. -- BenRG (talk) 23:19, 3 December 2009 (UTC)
 * I know, but as it happens, it documents a perfectly legal version of the function they are trying to write. As such, I was speculating that the difference involved might be the source of the problem. Seeing your post below, I'm guessing that the problem is completely different (possibly related to an error in the code using it, as opposed to the code declaring it), but the C++ compiler is doing its usual thing of providing useless error messages. &mdash;ShadowRanger (talk 23:22, 3 December 2009 (UTC)
 * The code is legal C++ and should work. And it does for me (Microsoft C++ version 15.00.21022.08 for x86). Is this part of a larger program? The error might be caused by an instantiation of the template rather than the template itself.
 * Incidentally, the semicolon after the close brace is not necessary and most people would consider it unidiomatic. It is legal, though. -- BenRG (talk) 23:15, 3 December 2009 (UTC)
 * I'm trying to make pclaf.obj using pclaf.cpp and pclaf.h at http://web.engr.oregonstate.edu/~budd/Books/oopintro2e/info/slides/chap19/slide1.htm by typing C:\>cl /c pclaf.cpp when I get the error. 71.161.61.41 (talk) 00:16, 4 December 2009 (UTC)
 * The problem (and I feel silly for not guessing this earlier) is that <tt>min</tt> and <tt>max</tt> are already defined as macros in an included header, so the compiler is seeing something like <tt>template <class T> T (T x > T y ? T x : T y) { return ... }</tt>. You can work around this by #undefing them before defining the templates. This is very old code, predating the ISO C++ standard, and would need a fair amount of cleanup to compile under recent versions of Microsoft C++ (which, unlike earlier versions, adhere pretty closely to the standard). -- BenRG (talk) 00:55, 4 December 2009 (UTC)
 * Thanks! 71.161.61.41 (talk) 00:58, 4 December 2009 (UTC)
 * Actually, the stray semicolon makes the program ill-formed, assuming that <tt>max</tt> is in the global namespace, and not a class member function template. A few compilers actually do diagnose it (most don't). <span style="font-family:monospace, monospace;">decltype (talk) 15:26, 4 December 2009 (UTC)
 * An aside: you only need to stick "typename" in when the compiler might guess wrong about whether a template-parameter-qualified name is a value or a type. "T::foo" could require "typename" because the compiler doesn't know if foo is like "typedef int foo;" or "static void foo {}" in class T.  It defaults to interpreting as a value, so if it's a type you'll need to say so with "typename T::foo".  Since your example doesn't contain any scope resolution operators (double colons) at all, you would never need "typename".  --Sean 16:08, 4 December 2009 (UTC)
 * I was surprised to see we didn't have a typename article, so I added one. --Sean 18:42, 4 December 2009 (UTC)

Why School and Library Computers Don't Suffer Often from Viruses
Why do the computers at schools and libraries don't suffer from viruses and malware as much as personal computers despite their high traffic? All of these public computers I have used erase everything stored on the computer when rebooted; could this be one of the reasons? Would setting up a personal computer like this (one that erases everything when shutdown) effectively eliminate all malware/viruses? 128.84.178.36 (talk) 20:04, 3 December 2009 (UTC)


 * Not all, no. If you get infected while logged in and then create some documents which also become infected, and send them via e-mail, or whatever, you'd still be propagating the malware.  The hard disk wipe does help a lot, though.  When the hard disk gets wiped at the end of a session, the malware gets wiped, too.  Trouble with doing this on your home PC would be that you could no longer change its configurations, add programs, store any documents locally.  Tempshill (talk) 20:10, 3 December 2009 (UTC)


 * Most malware these days isn't viruses, and mostly gets onto machines because someone let them (they fell for a scam and downloaded and ran something they shouldn't, or ran some program emailed to them under a deceptive guise). Machines set up as kiosks generally take some simple security measures with this in mind:
 * the ordinary user is an unprivileged account (and not an admin), and that account has its abilities heavily curtailed using the system security policy
 * the administrator is a professional, and is hopefully either very experienced or pretty well qualified
 * anti-virus and anti-spyware software are kept up to date
 * there usually isn't an email client (or it's disabled); users use webmail (which is generally less likely to transmit malware effectively to the machine).
 * So, for the merely incompetent users such a machine mostly gets, these sensible security measures are mostly sufficient. They do the blank-and-restore thing as an additional protection, and for protection against local users with malicious intent. For a single-user home machine, blank-and-restore is overkill, and if you're technical enough to set that up properly, you're technical enough not to get malware in the first place. -- Finlay McWalter • Talk 20:13, 3 December 2009 (UTC)


 * (ec) Another major reason they are probably not often infected is their rigorous policy that users do not log in with accounts that have administrator rights, so system files can't get infected. Most home and small business users just blindly use admin accounts all the time.  Tempshill (talk) 20:15, 3 December 2009 (UTC)
 * This is the crux of the difference between almost all home run computers, and computers that follow basic security guidelines. A current version of Windows with the latest security patches and a properly restricted user account is resistant to almost all viruses/malware.  Antivirus tools are little more than a safety net for haphazard users with too much privilege on the system.  It cannot be stressed enough, if you want to stop problems before they start, give users only enough privilege to log in and do exactly what they need to do.  Also, one more comment about the flush-at-reboot approach, this is obviously more final (especially if you boot up from unwritable CD or DVD and the hard drive is completely wiped) and it doesn't have to be impractical.  There are "live CD" boot images for many major operating systems that are easy to obtain, can be customized, and include a lot of features.  If the original question was asked with the intention of building a more robust computer for, say, a less computer fluent family member that won't require constant support to remove malware and correct configuration mixups, this may be a good approach to take. --Jmeden2000 (talk) 21:18, 3 December 2009 (UTC)


 * It's likely that most schools and libraries enjoy the services of one or several professional computer experts in a dedicated I.T. Department who are specifically employed to set up the computers and the network they're on securely, to install the best available security programs, to keep those programs fully up to date, and to actively monitor the network for problems, external attacks, etc: certainly that's the case in the commmercial companies with comparable equipment that I've worked at. 87.81.230.195 (talk) 00:35, 6 December 2009 (UTC)


 * When I was in college, my school's computer lab was staffed by students. They weren't the most knowledgeable people in the world.


 * Anyway, one of the lab computers we'd remote-desktop into got a really bad virus. We weren't administrators, and the system ran a fully-patched version of Windows Server 2003. It spread to three other computers. Removal of adminship will stop many viruses, but some will still be able to infect the computer using buffer overflows or by running under the system account at startup. I read about an exploit the other day that allows you to give yourself administrative privilages in a Vista machine. In later years, the lab used machines with Windows Vista and Windows Server 2008 and I never saw those machines become infected.--Drknkn (talk) 04:40, 6 December 2009 (UTC)


 * Some hints for using library and other restricted (so called "kiosk mode") setups:
 * (1) Use web-mail. (Duh.)
 * (2) Save favorites by copy/pasting them in a draft e-mail in your web-mail account.
 * (2b) You can then shift-click on the draft e-mail to get a full page of favorites. This works about as well as the normal favorites of the browser, and saves a lot of typing. Also you can mail your favorites to your other/backup e-mail accounts.
 * (3) Save image files you want to keep to a temp directory, and then attach them to a draft e-mail.
 * (3b) If saving temp files is not allowed, copy paste the whole page in a draft e-mail. This creates large e-mails, so use a web-mail service with lots of room. E.g. GMail.
 * I am actually writing this from a library computer. 195.35.160.133 (talk) 11:07, 8 December 2009 (UTC) Martin.

How can I tell what Bluetooth profile a USB dongle supports
I found a USB Bluetooth adapter behind my desk. How can I tell what class and profile it supports? There's no manufacturer name. --70.167.58.6 (talk) 20:13, 3 December 2009 (UTC)


 * If you plug it into a Linux machine and type <tt>lsusb</tt> it will show you the device's manufacturer and model number, give you something to put into Google to find info on the device. I think in Windows there's something similar in the DeviceManager (I'll get back to you on that). -- Finlay McWalter • Talk 20:17, 3 December 2009 (UTC)


 * Hmm, the Windows Device Manager (on 7 at least) has lots of USB properties for installed devices, but none reports the useful info that lsusb does. -- Finlay McWalter • Talk 20:29, 3 December 2009 (UTC)


 * Right-click any device and select properties. From this page select the details tab and then just scroll through the "Property" choices to find the information you want. The most useful ones are "Hardware Ids", "Bus relations" and "Device Instance Path" (which between them will tell you the manufacturer/model (if it's actually saved in the USB hardware) and Googling the Device ID will usually give you more information. <span style="font-size:14px;font-weight:bold;font-family:verdana;color:#800040;filter:Shadow(color=#000000,strength=1);height:15px;">ZX81  talk  20:51, 3 December 2009 (UTC)


 * So yeah, I'm using a Mac, so none of those options are available. Looking through Wikipedia, I found a photo of it. It's the exact one listed in the Bluetooth entry here  --70.167.58.6 (talk) 18:57, 4 December 2009 (UTC)

CSS "tabs"/indents
So I'm trying to make a page that looks something like this:

...and so on. Except, I'd like (for a variety of reasons), to do it without tables. Is there any way to get this kind of tab-like behavior? Indents don't seem to help much, because I can't separate the date and first line. Putting a margin-right on the date doesn't work because the lengths of any individual date can vary (e.g. "May 5" is fewer characters than "February 5" and so it looks incorrect). In Microsoft Word, this would be accomplished with tabs, but I don't see any obvious CSS equivalents.

Any thoughts? --Mr.98 (talk) 21:06, 3 December 2009 (UTC)


 * Make each date block a float:left div, each of the right "column" another float:left div, and do a clear:left between the row. -- Finlay McWalter • Talk 21:08, 3 December 2009 (UTC)


 * Like this:

January 1 Here is some information. January 2 Here is some more information. January 3 Here is yet more information. <br style="clear:left;">
 * Of course in practice you'd use CSS classes and a stylesheet (which I can't readily do on Wikipedia), making your html very simple.-- Finlay McWalter • Talk 21:12, 3 December 2009 (UTC)


 * That's clever, but it doesn't quite work for what I'm doing. One of the reasons I'm trying to avoid using a table is because I'm floating an image to the right of all this text (I know, how complicated can I make this? I swear it's not really that complex a layout visually!), and doing that with all of these other floats makes some weird things happen if the "some information" text is long (which it is in some cases)—it'll float the DIVs near the image in weird ways, like this:

January 1 Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. <br style="clear:left;"/> January 1 Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. <br style="clear:left;"/> January 1 Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. Here is some information. <br style="clear:left;"/>


 * Bleh. Any other ideas? :-/ --Mr.98 (talk) 21:23, 3 December 2009 (UTC)


 * OK! I figured it out. If I don't make the info floating, and instead just give it a margin-left of the right size, then the above works OK. Great! Thanks. --Mr.98 (talk) 22:04, 3 December 2009 (UTC)


 * I figured out a way with floating: http://mcwalter.org/unlinked/css_table/index.html (with stylesheet at http://mcwalter.org/unlinked/css_table/fin.css). -- Finlay McWalter • Talk 22:09, 3 December 2009 (UTC)

A definition list is good for data pairings. ¦ Reisio (talk) 05:27, 6 December 2009 (UTC)

domain cloaking, etc.
I run a page at a URL like http://myuniversity.edu/dept/. It is hosted by the university in question. Unfortunately the university's IT department is pretty backwards and the web options are pretty artificially constrained. I would like to have the hosting provided by a commercial host, like bluehost.com or dreamhost.com or whatever. The problem is, I need the URL to look the same (it still needs to be http://myuniversity.edu/dept/) and it needs to be seamless (that is, if you go to myuniversity.edu/dept/aspecificpage.html, it would have to work correctly and still appear to be at myuniversity.edu).

What specifically would the university IT people have to do in order to make the .edu domain point to the commercial host servers? How technically difficult would this be (especially considering that ONLY the "/dept/" directory would point to these servers—the rest of the university's pages would obviously still be hosted on university servers)? This is normally consider "domain cloaking", I believe, but I don't really know how it works on the back-end, or how it would work in the case of a single directory on a domain (and not the whole domain).

Does anyone know what this would involve? Goofy half-solutions (e.g. loading all the pages via AJAX or something) are not really acceptable possibilities. --Mr.98 (talk) 21:11, 3 December 2009 (UTC)
 * I can almost guarantee that your university will not change their DNS records for you which is required to do what you ask. The only other options are to do half-solutions that can be quite a pain and aren't really adequate.  Framing a site is a popular one in this case.  The cleanest way I can think of to show the other URL is legit is to have a link on your .edu page that says something like "due to technical restrictions, this department's homepage is located at x.url."  Of course, University policy can be pretty backwards even when this is taken into account so ask about this first.  You could also possibly get funds to set up your own server for the department where your options are less constrained by the normal IT department.  I am, of course, assuming that this is a site for a full department at the university. Caltsar (talk) 21:32, 3 December 2009 (UTC)


 * Framing and linking offsite won't work, no... it will look pretty cheesy and the whole goal is to not look cheesy. ;-)
 * If I know exactly what the DNS requirements are (e.g., what exactly they will need to do), I have a good chance of getting them to do it (not because they care a wit about me, but because the person I work for is pretty close with the head of their IT department and can get things pushed through if need be), but I will need to know exactly what is to be done (because the IT people will not know how to do it and won't want to bother with it otherwise). (Oh, if only the IT department didn't see the need to do everything in-house in the first place...) --Mr.98 (talk) 21:42, 3 December 2009 (UTC)
 * DNS doesn't have anything to do with directories, and can't be used to make just one directory appear on a different server. A Reverse proxy, if supported by the webserver on myuniversity.edu, can forward requests to /dept/ to a third server, making it seemless from the browser's point of view.82.75.185.247 (talk) 22:54, 3 December 2009 (UTC)


 * OK... blah. Reverse proxies look like a lot of work. Hmm. Dang. --Mr.98 (talk) 00:23, 4 December 2009 (UTC)


 * Another option is to actually host your own server at something like "yourserver.yourschool.edu", which is a lot easier than getting a subdirectory of another web host to point to your machine. Would that be an acceptable solution?  Nimur (talk) 22:38, 4 December 2009 (UTC)

Copying URLs to clipboard
I am using WinXP and IE8.

I would like to be able to highlight some text on a webpage (or just indicate the whole page) and have all the explicit URLs there copied to my clipboard while the rest of the text is ignored: is there any quick way to do this please rather than copying each URL individually? I only want to copy the visible URLs, I do not want to copy any weblinks where the text of the URL is not visible to the eye.

Second question, is there any quicker way to copy a visible URL to the clipboard other than highlighting it, right clicking, and choosing 'copy' from the menu? 89.242.105.246 (talk) 21:19, 3 December 2009 (UTC)


 * 2nd question: You don't have to highlight the URL; just right-click it.  Tempshill (talk) 21:41, 3 December 2009 (UTC)


 * 2nd question: Alt+D, Ctrl+C. --Andreas Rejbrand (talk) 22:31, 3 December 2009 (UTC)


 * Just to clarify, 89.242.105.246 seems interested in gathering addresses spelled out in text, but not interested in clickable links on the page. The only solutions I can think of involve writing a script or program.


 * Here's an AutoHotkey script that lets you select text, then press F12 to put just the URLs on the clipboard. (In this script, a URL is considered to be text that begins with http:// or https:// )


 * For example, with this script loaded, I clicked on this page, pressed Ctrl+A, F12, and the clipboard contained


 * --Bavi H (talk) 04:34, 4 December 2009 (UTC)

Thank you very much. I've installed AutoHotKey and run the script. However unfortunately on closer inspection many of the URLs I want to copy only make visible a shortened version of themselves on the webpage, putting in dots where parts of the URL has been skipped, while the underlying URL that I want to copy is complete. Is there any quick and easy way to mass-copy these URLs into the clipboard please? Thanks. 92.29.42.147 (talk) 13:59, 4 December 2009 (UTC)


 * I misunderstood and thought you wanted to gather mainly plain text addresses that aren't clickable links. But it sounds like you're interested in the addresses of clickable links, but only certain kinds.
 * Here's a way you can see the addresses of all the links on a page in Internet Explorer 8: Click on the Tools button, then click on Developer Tools. In the Developer Tools window, click on the View menu, then click Link Report. This will open a new tab that shows the address of every link on the page. This might make it easier to copy the addresses you want normally or with the AutoHotkey script above. --Bavi H (talk) 04:07, 5 December 2009 (UTC)

Thanks, I tried that, Unfortunately that produces lots of URLs, and it is difficult to identify the ones I want. Perhaps I should have a go at extending your script myself. Is there any way to identify the underlying URL (in the source code) that the visible URL refers to please? Or to filter all the URLs in the source code and only copy those that include a particular text string? Thabks again. 92.27.148.85 (talk) 12:55, 5 December 2009 (UTC)


 * AutoHotkey uses the plain-text version of the clipboard, so it can't access the addresses of links you select on a normal webpage. However, if you select text on the Link Report list, then the text of the link (which AutoHotkey can access) is the same as the address of the link. So you can select all the text on the Link Report, then run the script to get a list of all the addresses on to the clipboard. Here are some ideas to make that list more useful:
 * If you sort the list, that might make it easier to identify the addresses you want. Add the line Sort, out right before Clipboard := out
 * Read the AutoHotkey help file pages "RegExMatch" and "RegEx Quick Reference" to learn how to change the filter in the script.
 * --Bavi H (talk) 17:48, 5 December 2009 (UTC)


 * Also, here's a completely different idea. I know installing another web browser is a large step to take. But in Opera, you can go to Tools, Links to see a list of links on the page. (It looks like this: opera links.gif.) You can click on the column headers to sort (or unsort) the list by name or address, and you can type in the little search box to filter the list. Then you can Shift+click or Ctrl+click to select multiple links you want, then use Copy (Ctrl+C; Edit, Copy; or right-click, Copy) to copy just the addresses. --Bavi H (talk) 18:24, 5 December 2009 (UTC)

Sensitive Data What to do ?
Well, is there some way I can hide my important .doc, .txt and picture files. Of course I know about "hide" command but anyone can tick the "show hidden files" option and see them. I mean no one can see them no matter how hard they try. Or a part of my harddisk becomes inacessible to anyone but me...?

Jon Ascton   (talk)  22:24, 3 December 2009 (UTC)
 * You could create a TrueCrypt volume on your disk. To anyone else, this will just be a single file containing data that appear to be random. If you click on the file, you will be prompted for a password to mount the disk. After having given the password, the volume looks just like a second hard disk for the remainder of the session. You could save and edit your sensitive data there. The safety of the data, will of course depend on the strength of the password you choose. --NorwegianBluetalk 22:36, 3 December 2009 (UTC)


 * Bruce Schneier's advice in this regard is here. -- Finlay McWalter • Talk 22:58, 3 December 2009 (UTC)

Sony 16GB S Series Walkman Video MP3 Player -- Windows 2000 support
I may soon get a "Sony 16GB S Series Walkman Video MP3 Player" (model NWZ-S545). However, my operating system is Windows 2000, which Sony's "Content Transfer" program claims to not support. I'd personally prefer not to use Content Transfer anyway, but I want to be able to use the player. A review online says that when you use a different operating system, it accesses the player as a USB mass storage device, which I assume I'd greatly prefer over having to use Sony's middle-man program on my old, slow computer.

Does anyone have this player, and use or have access to Windows 2000? If so, can you tell me whether I'll be able to properly get my files onto it (hopefully just as a regular removable storage device)? For the record, I have nothing that is DRM-protected, so that's no issue. Also, if I have to use MediaMonkey to transfer, I can do so. ---4.251.126.53 (talk) 22:39, 3 December 2009 (UTC)
 * If the MP3 player is USB Mass storage device compliant then it means that when you plug it into a Windows 2000 computer (or newer) then it will operate like a USB pen drive, so you will we able to simply copy over music files in the same way you would copy over other files to a USB pen drive. Depending on the player's configuration you may have to copy them to a specific folder on the player, but otherwise you won't need any software to copy of your files at all. Rjwilmsi  20:17, 5 December 2009 (UTC)


 * The review that I read said that when using Windows, you have to access the device in MTP mode. I'm not too familiar with that, as I've never gone above Windows 2000. One of their complaints in the end summary was that they were forced to use MTP. However, they said that in other operating systems it uses UMS. The reason I'm confused is because W2K doesn't even have an MTP mode, so I'd assume that the computer would just treat it as mass storage by default... but I don't know. --4.251.126.205 (talk) 21:08, 5 December 2009 (UTC)