Wikipedia:Reference desk/Archives/Computing/2010 May 4

= May 4 =

Wikipedia date being updated
Every day the English Wikipedia Main Page presents a different FA, a different FP, different On this day and Did you know sections. So I can see the new content of the page, it's enough to click on the reload button of my browser exactly at 00:00:00 UTC. However, reloading doesn't work so momentarily for some other Wikipedias, especially for some smaller ones. Hitting Ctrl+F5 sometimes doesn't work either; I therefore have to purge the Main Pages to be able to view the content corresponding to the current date. I'm just curious - why so? If that matters, the browser I usually use is Mozilla Firefox. Thanks. --Магьосник (talk) 00:22, 4 May 2010 (UTC)
 * My guess would be because they're small and thus don't have as much people working on them, let alone the Main Page. —  Jeremy  ( v^_^v Dittobori ) 00:36, 4 May 2010 (UTC)
 * This has been done to change automatically, hasn't it? Is the automatical updating of something like this:  affected by the number of people working? I will try to rephrase what I'm wondering about. If I visit the English WP Main Page for the first time today, say, at noon, then it will always show the current date. If I do the same with, say, the Icelandic WP, it will often show the previous date. And I'll be able to correct this just in a second by purging the page. That seemed odd to me, and I decided to ask about it here. --Магьосник (talk) 00:56, 4 May 2010 (UTC)
 * I don't really know, but I wouldn't be surprised if there was a bot that purged the Main Page a few seconds after midnight UTC. It would take nothing but a trivial one-line cron entry running on the toolserver.  (Hmm, OK, maybe a few more lines, since it'd have to either log in or deal with the confirmation button that action=purge gives to anon users.)  It would take me all of fifteen minutes to set up, so it seems quite likely that someone has else already done it.  Presumably whoever it was wasn't Icelandic, though.  —Ilmari Karonen (talk) 12:14, 4 May 2010 (UTC)

How does RAID1 behave in the face of a single disk corruption?
After having a hard drive fail on me, I have considered using a RAID1 setup for my replacement, but there is one scenario that bothers me. RAID1 can obviously save your data if one drive is completely broken (i.e.: can't be read), but what happens if the drive still operates while returning incorrect data? What if the data on one of the drives becomes corrupt? How would the RAID1 controller/software know which drive is returning the correct data? If it can't tell this, what happens? Will there be a risk of the corrupted data propagating? --35.11.183.86 (talk) 02:37, 4 May 2010 (UTC)
 * Firstly, while not really answering the question, RAID 1 is not a replacement for good backup practices. There are plenty of cases when RAID 1 won't protect against data loss, e.g. file system corruption, accidential or malicious deletion (including malware), simultaneous failure of both hard drives (e.g. due to catastrophic power supply failure), or near enough (e.g. if they are from the same batch they can sometimes fail at around the same time and if you are in a server or other situation where you can't afford downtime so keep running until you get a replacement for the broken disk, such a situation is really one of key reasons for RAID), some sorts of controller failure... If you are thinking of using RAID 1 as a substitute for good backup practices, I strongly recommend against it. In terms of your question, if the drive knows the data is corrupt, then that shouldn't be a problem. Even if it doesn't you may be able to manually work out which drive is corrupt, by comparing the differences and trying to work out which one is correct (if it's file system information, or a file you have a checksum of, or a file with builtin checksum verification like an archive it should be easy). In terms of how the controller handles it, I don't really know much about RAID 1 but if I'm not mistaken you can set it so it does alternate read requests to speed up reads. If you do this then clearly the controller/software is NOT going to know the data being returned is different, since it's not reading the same data twice. I presume you can set it up with some hardware/software so it doesn't do this and instead reads both and then compares, in such a case, I expect you receive some sort of error message about inconsistent data and you should then shut down and backup as much as you can of both disks and then work out what the problem is. But for good measure, let me repeat again that using RAID 1 doesn't mean you should stop backuping up data (if you aren't you may want to consider starting and if you can't afford to do both, it's probably better to backup rather then set up a RAID 1 array). Nil Einne (talk) 04:35, 4 May 2010 (UTC)


 * From personal experience: If two disks are run in RAID1 mode and have identical directories and files, yet different file contents for the same file (because of an error on the disk or because one of the disks was accidentally mounted as an ordinary, non-RAID disk, and then re-joined into the RAID), you'll end up with garbage, as the controller/driver picks one chunk from disk 1 and the other from disk 2. Obviously, with two disks, there is no room for checksumming.  If that's what you want, you will have to use a RAID5, which requires at least 3 disks, one of them keeping the checksum. -- 78.43.60.58 (talk) 13:23, 4 May 2010 (UTC)
 * I don't see how RAID5 improves matters. In either RAID1 or RAID5, the disks aren't usually checked for consistency for performance reasons, and there would be no way to automatically tell which disk was at fault if you did find an inconsistency. You need at least three parity blocks to correct one error, and RAID5 only has one. -- BenRG (talk) 17:43, 4 May 2010 (UTC)
 * Gah, you're right, RAID5 only keeps the checksum data for restore purposes in case of a detected disk failure. Would be nice if one could enable check-on-read, though. ;-) -- 78.43.60.58 (talk) 17:53, 5 May 2010 (UTC)
 * High-end storage arrays (the kind whose cost is on the order of my salary) often provide a background "parity scrub" feature that periodically or continuously scans the disks checking for parity errors. But you're right, "RAID5 does not provide a means to locate or correct an error detected in the above manner."  -- Coneslayer (talk) 19:46, 4 May 2010 (UTC)

Why is steam download so hard to find?
I realize this may be a subjective question, but I recently was trying to download Valve's Steam application to my new Windows 7 partition. In the past the only thing you had to to do was go to Steam's web site and there was a big download icon right in front of your face (https://support.steampowered.com/kb_article.php?p_faqid=144).

Now, whenever you go their web site there is no download link on the front page. It took me a few minutes of searching to find out you have to click on the About link on their front page to get the link to download the file.

Why would they make this so obtuse? Who is going to click on the About link on a web site? Usually that's where you find out about a company's philosophy, goals, etc.; I usually could care less. Why wouldn't they want people downloading their application the second they went to their web site? I don't understand the thinking behind this. —Preceding unsigned comment added by 70.245.111.56 (talk) 06:18, 4 May 2010 (UTC)


 * Yeah, no kidding. I've never understood why it was hard to find before (it was not a very big or noticeable button), and now it's even worse! Man. Indeterminate (talk) 09:44, 4 May 2010 (UTC)
 * I think you mean that you couldn't care less. If you could care less, then you care at least a little bit.  I know, this isn't the language desk...  Dismas |(talk) 10:12, 4 May 2010 (UTC)


 * Just click "About" on the menu, there's a large green "Install Steam Now" link on that page. ZX81  talk  10:18, 4 May 2010 (UTC)


 * Well they do show you a big download link after you've purchased a game... Maybe they figured that you probably don't need Steam if you don't have any Steam games? --antilivedT 10:26, 4 May 2010 (UTC)

FORTRAN Programming (Lahey Compiler)
I am trying to run one old fortran program (1980's). I compiled it using Lahey Compiler and i removed all errors and warnings. It is not running properly, if i give some print or write commands at some particular locations in some subroutines, it starts running. I am not able to understand without making any change in the code how is it running just due to some print or write commands and the results are still some what away from expected.203.199.205.25 (talk) 04:25, 5 May 2010 (UTC) —Preceding unsigned comment added by 203.199.205.25 (talk) 07:16, 4 May 2010 (UTC)


 * It is safe to say that there is no way we can possibly answer this question in its present form. Consider reading our guide on how to ask a software question.  If you inherited legacy source-code, and then modified it, and now it doesn't work, there literally infinite numbers of possibilities for what is going wrong.  Consider reading our debugging article to gain some insights into the methodical process of narrowing down what is broken.  If you can provide a simple, short code-fragment that demonstrates the problem, we can start helping you.  Additionally, the methodical process of identifying which portion of your code is actually the broken part will also help you figure the problem out.  Try to identify the last portion of the code which is running properly.  By definition, the bug immediately follows that code section. Nimur (talk) 10:59, 4 May 2010 (UTC)


 * Actually, I've encountered exactly this type of bug before, where adding a print changes whether the program runs. This is usually the sign of a memory leak, as when an array or character string is filled beyond the declared length, and thus happily overwrites whatever follows in memory.  This may or may not cause an error, depending on what follows in memory, and adding print statements can affect that.  One way to debug this is to temporarily declare each array or string to be huge (obviously within memory limits), one at a time, and see if that fixes the problem.  When you find the problem array or string, track it through the program to try to find the place where it overflows.  Note that you often don't get an error right where the overflow occurs, but later, when the program tries to use whatever was in that chunk of memory.  This makes such bugs difficult to track down, as does their intermittent nature.  StuRat (talk) 18:41, 4 May 2010 (UTC)


 * I'm not familiar with the Lahey Compiler, but most compiler have a variety of switches to turn on stricter standards compliance, array bounds checking, and produce additional listings of data disposition, array usage, machine code, etc. Failing that, there might be various debug options to allow you to step through the program ones line or instruction at a time and display the status of all kind s of things as you go.  Astronaut (talk) 19:03, 4 May 2010 (UTC)


 * Manuals for current versions of the Lahey Compiler are available [here]. According to the LF Fortran Professional manual you can get the most comprehensive automatic checking by compiling all libraries and program units with the -chkglobal option.  If you depend on pre-compiled libraries, you can try the -chk and -trace options.  The specific section of the manual is found under LF Fortran 95 -> Fortran 95 Reference -> LF95 Compiler and Linker Options.  It's definitely worth a try!  130.188.8.10 (talk) 10:49, 7 May 2010 (UTC)


 * There's a technique I sometimes use to track down a bug. Here's a chunk of code with an obvious bug:

PROGRAM WB CHARACTER*10 SUSPECT DO I = 1,99999 SUSPECT(I:I) = "X" ENDDO END


 * But let's imagine it wasn't so obvious (perhaps the value of I is determined by some complex formula, rather than just incrementing in a loop). I add a junk string before and after the SUSPECT string, and check for when either gets overwritten.  (Depending on the compiler, either junk string may follow the SUSPECT array in memory, possibly with a short gap in between.)

PROGRAM WB      character*10 junk1 CHARACTER*10 SUSPECT character*10 junk2 junk1  = " " ! Blank out both strings. junk2  = " " DO I = 1,99999 SUSPECT(I:I) = "X" if ((junk1 .ne. " ") .or. (junk2 .ne. " ")) then print *,"Memory overflow detected at I = ",I print *,"junk1 = >",junk1,"<" print *,"junk2 = >",junk2,"<" stop endif ENDDO END


 * This produced the following output on my PC:

Memory overflow detected at I = 129 junk1 = >X        < junk2 = >         <


 * So, apparently my compiler allocated 128 bytes to the SUSPECT array, then allocated junk1 right after that. StuRat (talk) 19:32, 4 May 2010 (UTC)


 * Sorry as i had forgot to put tildes in my question, i have corrected it. Actually my program has several subroutines, in one of the subroutines the program was getting stucked. i started printing all the variables and i found some of them as NaN. i located the problem in this subroutine and i gave a print command for variable 'TEDTRB(NRB)', it was NaN but when i gave print command either for 'ALFA' or 'TSTART', it starts running. as suggested by user StuRat, i will change the dimensions of all the arrays to a big number and see what happens. as suggested, i am putting the subroutine below. if anything else is to be given, please tell me.203.199.205.25 (talk) 07:16, 5 May 2010 (UTC)

203.199.205.25 (talk) 07:16, 5 May 2010 (UTC)


 * I have rolled-up the source code. Nimur (talk) 13:55, 5 May 2010 (UTC)


 * Try checking the value of NRB. If it is less than 1 or greater than 50, it will access an invalid element of the TEDTRB array.  It might also be worth checking other declarations of TEDTRB throughout the whole program.  If someone has changed the parameter used to set the size of the TEDTRB (IRB=50 in the above subroutine), but failed to change it here, that could cause all kinds of havoc with the layout of COMMON/TEMPA1/.
 * On another note, the code above exhibits many of the aspects of programming that make people think (incorrectly IMHO) that Fortran is a bad language to program in. Subject to the compiler standards that are imposed, it could improved with:
 * Reorganisation to avoid the use of the GOTO statement
 * Use longer, more descriptive variable names
 * Move the commons and the associated size parameters into one or more separate files which are included where needed using the INCLUDE statement. This makes future maintenance easier and reduces the likelyhood of errors like I described above
 * Use of spacing and lowercase to make it more readable
 * Add comments explaining what it is meant to do and what all the variables do
 * Astronaut (talk) 15:16, 5 May 2010 (UTC)
 * Sorry friends, its not working. i checked with increasing the dimensions of the arrays as well as i checked the value of NRB, its 21. Any further suggestions?203.199.205.25 (talk) 11:15, 6 May 2010 (UTC)
 * NaN's (and floating point overflow/underflow) often point to an unintentional mixture of integer and floating point arithmetic. Some compilers are more forgiving than others. Zoonoses (talk) 14:42, 7 May 2010 (UTC)


 * One of the variables in my program is DTE(L) = 1.E75,  when i compile, i get the error message as "Real or complex values overflows", i searched it on internet and found that writing it as " 1.E75_8" removes the error, whether this is creating any problem? plz suggest me.203.199.205.25 (talk) 03:58, 8 May 2010 (UTC)
 * 1e75 is a very large number. The Observable universe article suggests the number of atoms in the observable universe is something like 8e80.  1e75 won't fit in an R*04 (range is about -1e38 to 1e37, more or less depending on how it's implemented).  Defining your variable as an R*08 may allow it to compile cleanly, but why is it so large?  Zoonoses (talk) 12:15, 9 May 2010 (UTC)

android browsers
I've put two android browsers on my android phone. The one is the standard android browser that comes w/ the phone. I also installed Skyfire. I want one set of bookmarks on the one browser, lets say the 'work' bookmarks, and a different set on the other browser, let's say the 'home' bookmarks. But they both seem to know what the other does, and bookmark and remember on both browsers what is accessed on the other. How do I separate them so that they operate discreetly (so to say). —Preceding unsigned comment added by 80.101.134.43 (talk) 09:13, 4 May 2010 (UTC)

Phishing - what is going on here?
Hi. I recently had a virus on my PC which had performed some kind of browser hijack. Once I cleaned the virus IE and Opera would no longer connect to the internet and gave a proxy related error. Of course, that was because the virus had installed an executable that was acting as a local proxy, and when it was removed the proxy was no longer available. So, I went into IE and Opera and set the connection settings back to use no proxy - which fixed the problem. During this time Firefox seemed unaffected and does not seem to have been configured to point to a proxy. I have run several complete scans (including heuristics) using F-Secure and the machine seems clean. Hwoever, this morning when I went to log into my online banking using Firefox, I got a very suspicious login screen:

http://www.flickr.com/photos/49893252@N04/4577352131/sizes/o/

which differs substantially from the usual login screen:

http://www.flickr.com/photos/49893252@N04/4577984292/sizes/o/

As you can see, the suspicious screen asks for all kinds of card details that an online banking screen would never ask for. I called the bank and they confirmed that the screen is fake, but were not able to explain what was going on.

Looking a bit deeper, it seems that the site itself is valid and has a valid certificate:

http://www.flickr.com/photos/49893252@N04/4577984360/sizes/o/

which appears to match the certificate for the site displayed by Opera:

http://www.flickr.com/photos/49893252@N04/4577352431/sizes/o/

I have tried to connect to the same site via Firefox on a different machine, and I get the non-dodgy login, but exactly the same certficate (with the same fingerprints) as dodgy login page on the original machine.

I wonder if anybody could please advise what is going on here, and what I can best do about it.

Any help would be really appreciated. Thanks GoldenStick (talk) 10:06, 4 May 2010 (UTC)


 * I like the comic touch in the fake screen: (i) asking you for your PIN, and (ii) warning you about any demand for your PIN.


 * If matters are like this, I'd (i) remove the machine from any LAN; (ii) boot off Clonezilla to use it to copy the partition or logical drive where you keep your own files to an external drive (skip this stage if you're one of those unusual people who actually backs stuff up frequently); (iii) boot off some other Linux CD to repartition the drive and wipe out the MBR (to make changes as sweeping and deep as are possible); (iv) install your preferred OS afresh (and if I were you, my preferences might be affected by this experience); only then reconnect to the wider world. -- Hoary (talk) 11:02, 4 May 2010 (UTC)


 * My guess would be that the virus installed a Firefox extension that modifies the page, and that for some reason F-Secure did not notice or recognize the extension as malicious. You could try simply looking at the list of installed extensions (under Tools → Add-ons) and seeing if you can spot anything suspicious there that you don't remember installing.  Failing that, you could also peek into Firefox's extensions folder and look for any suspicious entries there.  If you do find something, you can send it to F-Secure for analysis, so that they'll hopefully add it to their malware database.  —Ilmari Karonen (talk) 11:52, 4 May 2010 (UTC)


 * That is a beautiful piece of malware! It looks like they intercept your bank login and add HTML components (this might be done by redirecting you to a malicious proxy server, or a Firefox add-on similar to Greasemonkey - that appends some pretty smart javascript to your bank login webpage).  That would circumvent the fingerprinting process - the page wouldn't actually be modified until it hits the client-side.  It is evident that this is a hijack, and that you should not submit any financial information until you can validate that your machine is clear.  Evidently, they are performing a man in the middle attack of sorts - allowing you to connect to your bank, but appending these extra fields to steal your ATM details.  What amazes me most is the delicacy of programming such a precise piece of work, and then flubbing up the strings: "Password and Pin, like that: Password-PIN" does not sound like instructions from an actual bank website!  It sounds like an ESL hacker!  *sigh*... Your machine is compromised.  If you are not capable of guaranteeing a complete removal of the malware, (which is very hard), a full reinstallation of the operating system is in order.  Nimur (talk) 13:11, 4 May 2010 (UTC)


 * It's not really that amazing. I could whip up a quick Greasemonkey script to do something like that in a few hours at most — modifying web pages locally is what Greasemonkey is designed for — and then run it through a GM-to-XPI compiler to turn it into a standalone extension.  The only slightly tricky part would be installing it without the user noticing, but I'm pretty sure any halfway competent malware writer could come up with some trick to either suppress the new extension notification or make it look innocuous, given that they already have control of your computer.  —Ilmari Karonen (talk) 13:55, 4 May 2010 (UTC)


 * Since I don't think you can guarantee you've rid your system of the malware, I would back it up, treat the backup like deadly poison, reformat, and start from scratch. You seem very technically adept, but Reference desk/Computing/Viruses is a FAQ on malware removal if you're interested.  Comet Tuttle (talk) 17:06, 4 May 2010 (UTC)
 * Try running Norton power eraser and/or Malwarebytes' Anti-Malware --Tyw7 (☎ Contact me! • Contributions)   Changing the world one edit at a time! 15:15, 8 May 2010 (UTC)

My earphones/earbuds dont work
I only had them for a coupple of weeks, so they were relatively new. Today they were working one minute, and now only the left side is working properly. The jack/plug and phones/buds are not damaged and the wires are not cut. Can anyone tell me what is wrong and how i can fix it? Thanks.

Also, does anyone know where there is a good translating program that translates modern english into old english, like Shakespearian? Google Translate does not have that and I couldn't find any websites that actually have such programs. 64.75.158.193 (talk) 11:48, 4 May 2010 (UTC)
 * Might sound overly obvious, but I'd make sure the jack for the earbuds is plugged in completely. If it is, and you're still hearing only out of the left side, sounds like a failure in the earbuds themselves.  If they didn't get damaged somehow (caught on something and pulled too hard, etc.) I'd probably return them to the store for an exchange.  I'd also suggest that you repost your question about the language translator at the Language reference desk. Kingsfold (talk) 12:06, 4 May 2010 (UTC)


 * Alright, thanks. I'll do that. 64.75.158.193 (talk) 12:51, 5 May 2010 (UTC)


 * I agree with Kingsfold. If your headphone is fairly new (under 1 year) then it is covered by warantee. Either return to the store you bought it from or ring up the manufacturer. --Tyw7 (☎ Contact me! • Contributions)   Changing the world one edit at a time! 15:13, 8 May 2010 (UTC)

Spellcheck in Outlook Express
The spellcheck in my Outlook Express seems to have a French dictionary not an English one. Is there a setting somewhere that I can change? DuncanHill (talk) 16:50, 4 May 2010 (UTC)
 * Hmm, I've found the Options/spelling bit, and it only offers me French - is there a download of an English dictionary somewhere? DuncanHill (talk) 16:52, 4 May 2010 (UTC)
 * Have you recently installed Office 2007? It seems Office 2007 changed the format of the spellchecker dictionary for all languages except French.  The same dictionary is used by Outlook Express, but since only the French one is in the old format, only French appears as a valid spellcheck language in Outlook Express 6 after upgrading to Office 2007.  It seems Microsoft doesn't see any need to issue an update for Outlook Express because you will all be upgrading to Vista and using Windows Mail pretty soon, right?  Three years later and with many people still on XP & OE6, people are still complaining about this.  See here and particularly the bit that says: "If you have Outlook Express 6.0 and installed Microsoft Office 2007 but French is the only available spell check language, you need to use a third-party spell checker. For more information, see Spell checking issues with Outlook Express 6.0".  This KB article is also very helpful.  Astronaut (talk) 18:54, 4 May 2010 (UTC)


 * Somewhere I read that you have to install Microsoft Live to get the English installed. 71.100.1.71 (talk) 19:29, 4 May 2010 (UTC)


 * Thanks, and yes I have installed Office 2007. Aren't Microsoft wonderful. DuncanHill (talk) 19:34, 4 May 2010 (UTC)
 * Yes, it are. Sorry, couldn't resist. Kingsfold (talk) 11:01, 5 May 2010 (UTC)
 * No they isn't. DuncanHill (talk) 23:10, 6 May 2010 (UTC)

My monitor has the shakes
My monitor (a Sony GDM-FW900 CRT) has developed a vibration recently, especially at the sides and the bottom of the screen (at the top in the centre it's ok). I changed the resolution (from 1600 x 1024 to 1440 x 900), but that didn't help. And if I lower the frequency (from 90 Hz to 80, 70 or 60 Hz) that only makes the vibration worse (the lower the frequency the worse it gets). Of course it could simply be that the monitor is dying, but although it's about 8 years old, it was rather expensive, so it should live longer than that (and I'd like it to last as long as possible, until sizeable LCDs that can double as a tv become more affordable). So any ideas what I might try? DirkvdM (talk) 18:05, 4 May 2010 (UTC)


 * How long has it been doing it and is it doing it all the time? My old CRT got the shakes maybe two or three times, and it would last a couple of days at most and then seemingly clean itself up. I never figured out the cause, but it was obviously transitory in my case. First thing I would do is make sure all connections are clean and secure - unplug the data cable and the power cable, make sure there's no crud in there, and then make sure they're both securely re-attached. Matt Deres (talk) 18:23, 4 May 2010 (UTC)


 * (ec) I used to see wobbles in my early Macintosh models when there was a problem with the flyback transformer, but this is not a user-serviceable part and you'd have to have a technician look at it. Comet Tuttle (talk) 18:25, 4 May 2010 (UTC)


 * (ec) Have the cables moved in any way? Signal bleed can cause video shake. I personally get more expensive cables simply to get better signal insulation. However, it is possible that the monitor itself is going out. The first thing I always check is the flyback circuit. Scanning online, it looks like it is in the back left (when facing the screen). Listen to it closely (without removing the case). Do you hear gurgling? If so, it is going to die and it will most likely be cheaper to buy a new one than replace the broken parts. I hope fiddling with the cables improves it though. I've always hated that it costs more to repair a good monitor than it costs to replace one. -- k a i n a w &trade; 18:28, 4 May 2010 (UTC)


 * I found that this is more of a problem with my CRT when showing brighter pics. I therefore set my background color, both on my desktop and here in Wikipedia, to black, and also turned the brightness down, and now it's fine. StuRat (talk) 18:31, 4 May 2010 (UTC)

Ah, thanks Matt! I thought the connections should not be a problem because I had them screwed tight, but wriggling the connection to the monitor (required a bit of 'caving' to reach it) solved the problem. As a friend of mine says, always start with checking the hardware-connections. I should listen to him more often. :) Anyway, I'm glad it turned out to be something as simple as this. Thanks! DirkvdM (talk) 19:10, 4 May 2010 (UTC)

Oh, a bit late, but I just realise that the connector was not really the cause. Because the monitor got the shakes again, I tried wriggling the connector again, but to no avail. So I tried to get closer and moved a power outlet splitter (what's that called?). And then I realised therein lay the problem. In that splitter I had two transformers, and apparently they were the cause, because I can adjust the shakes by moving the transformers. So I moved them well away. 50 cm appeared to be enough. DirkvdM (talk) 13:25, 16 May 2010 (UTC)

Creating a clickable icon for Command Prompt commands
Any suggestions on how to create a desktop icon that will run several command prompt commands? So the user never has to deal with the command line stuff? Just double click an icon to run several commands (running several Ruby scripts). --70.167.58.6 (talk) 18:28, 4 May 2010 (UTC)
 * On Windows? I (personally) have, in the past, created .bat files (a text file containing the commands, one on each line, AFAIK), which you can then create desktop shortcuts to, as I have. Hope that helps, - Jarry1250 [Humorous? Discuss.] 18:31, 4 May 2010 (UTC)
 * What do you want clicking this new icon of yours to do? Please, tell us more. --peterblaise (talk) 10:21, 6 May 2010 (UTC)
 * My vote's on the batch processing file as well. You can do a lot of stuff from inside those. I've seen some killer stuff done by bats. Cheers, Ouro (blah blah) 13:55, 6 May 2010 (UTC)

simultaneous two-way communication on telephone
I'm pretty sure that traditional land-line telephones allow both parties on a call to speak and be heard at the same time. With my cheapo mobile phone and service (Virgin mobile) this is not the case; if the two people try to talk at once only one person can be heard at time, which I find very annoying. My question is, is this always the case with mobile phones? If not, what is most likely the source of the problem, the cheap phone or the cheap service? ike9898 (talk) 18:55, 4 May 2010 (UTC)


 * The question is asking about duplex (telecommunications). Most handsets and networks are actually fully-duplexed (and support simultaneous transmission and reception) - though some very uncommon units and networks are only half-duplex (either transmit or receive, but not simultaneously).  However, the situation for modern digital phones (including land-line analog POTS phones, which connect to a digital network through a compatibility layer by the provider) is a bit more complicated.  Even if duplex mode is supported, advanced digital squelch, echo cancellation, noise suppression, and automatic gain control may all be operating on the signal as well.  These can be installed as software or hardware inside the handset, or may be realtime processing steps applied to data as it transits the telecom network.  This is, technically, a "feature" and not a "bug" - although it is annoying, because it is somewhat unintuitive - especially if you want duplexing behavior.  Any combination of these technologies may effectively squelch (mute) your signal, with the intention of improving your ability to hear the other party (especially at low-bitrates or in ambient-noise environments).  In practice, this interferes with the ability to fully duplex, even if the handset and network permit it.  Nimur (talk) 19:11, 4 May 2010 (UTC)


 * Thanks. The part that I still don't understand: If I were using a different commonplace consumer phone and/or a different commercial network, would I likely experience this problem less? ike9898 (talk) 19:49, 4 May 2010 (UTC)


 * Not really sure what Nimur is talking about, TBH. The fact is that mobile phone systems (the network and handset) are designed to support both parties talking at once.  However, owing to the time taken to code the speech into the digital radio communication, there is a delay between one party speaking and the other hearing. That can mean that it appears that both are trying to speak at once, and both then stop.  So talking simultaneously on a mobile phone is quite possible, but what is heard will be somewhat different, owing to the delay.  --Phil Holmes (talk) 20:26, 4 May 2010 (UTC)


 * Here is a patent for POTS with squelch (1974); here is a patent for a thresholding squelch for wireless phone (1978); here is a squelch patent for TDMA (old-style digital mobile telephone) (1999), here is Application Note AN014 from Texas Instruments describing software squelch for CDMA (modern American-style mobile phone); here is a book, Data over Radio, (1992), discussing CTCSS and more advanced digital squelch for packet-switched mobile telephoney, and here is a patent for squelch in GSM. As you can see, every major mobile telephony platform (analog, digital, TDMA, CDMA, WDCMA, GSM, 3G, whatever), all use a digital squelch or conversation-suppressor, even though they are capable of bidirectional transmission.  This is a design feature; it improves the effective signal to noise ratio of the audio, and allows lossier compression (ergo, lower bandwidth).  You can, if you so desire, describe squelch as an extremely low frequency lossy compressor: it turns low volume (rather, un-keyed) conversation into "zero data transmitted," even though the telephone supports duplex mode.  Nowadays, because squelch is often software-controlled, the mobile phone may opt to permit only one party to transmit, and mute the other, by "estimating" which one of you is talking (by simple analysis of volume threshold, or more sophisticated frequency or pattern detections).  If the system always opts to squelch one of you, but you are both talking, you will confuse the software squelch or cause it to thrash between the two parties - hence the strange digital noise you may hear.  Nimur (talk) 21:31, 4 May 2010 (UTC)


 * I see lots of patents describing how squelch could be implemented in radio transmissions. I don't see evidence that it routinely is.  All current mobile/cellular systems use circuit switched technology - therefore they reserve bandwidth for the call, for the duration of the call.  Implementing silence suppression would not bring benefit to the network operator, and I therefore don't believe it is implemented on most cellular systems. --Phil Holmes (talk) 07:59, 5 May 2010 (UTC)
 * Valid nitpick - I can't really say more than "WP:OR: accept on faith that it is widespread," because commercial networks and handset manufacturers are a bit reluctant to publish their compression schemes, device schematics, and circuit diagrams. Even an open-source platform like Android OS does not expose these proprietary communications-layer hardware details.  Nimur (talk) 14:00, 5 May 2010 (UTC)
 * Actually, the compression schemes are standardised across the world. For GSM, ETSI ensures that everyone uses the same Codecs, compression, etc., etc.  And, whilst I know that telling someone something you actually know could be described as OR, it surely is valid to do so on the ref desk.  I'm speaking from something like 35 years in telecommunications.  FWIW this morning I also spoke to an ex-colleague who represented O2 at the standards fora and he confirmed my statement that silence suppression/squelch is not used on GSM and W-CDMA 3G mobile networks.  --Phil Holmes (talk) 16:21, 5 May 2010 (UTC)
 * I guess I defer to your expertise in the field; I know it is technically feasible to implement these features. However, if your experience and ground-truth are correct, then such suppression schemes are not causing the interference that the OP refers to. The OP's interference is due to some other cause.  Nimur (talk) 15:11, 6 May 2010 (UTC)

My question really isn't that technical. Restated: Do you think pretty much everyone experiences what I experience, and to pretty much the same extent? Basically, it just bugs the shit out of me and I want to know if switching phones or providers would do any good. ike9898 (talk) 13:09, 5 May 2010 (UTC)
 * There is no reason that you can't both talk at once and be heard (with the proviso I gave above about delays meaning you're not heard at exactly the time you speak). If you can't do this, I would be surprised if it's the service provider and would suspect the phone is playing up.  --Phil Holmes (talk) 13:36, 5 May 2010 (UTC)
 * Every cell phone I've used has supported full duplex with no noticeable problems. Just like a land-line phone.  Talking on the phone would be very irritating if I couldn't occasionally interrupt the other person.
 * (Some phones have a 'walky-talky' mode that's cheaper, doesn't use your minutes, whatever, but I'm not sure I've heard of one that didn't have a regular mode.) APL (talk) 03:26, 6 May 2010 (UTC)

Visual Basic V6 versus VB.NET
... with the exception of now needing to find a distributed or cloud system on which to run the code. Going by Windows Azure it turns out that cloud computing comes down to little more than a Time-Sharing network of the 1960's with costs not low enough for this application.

While you might expect differences in programming language versions it never ceases to amaze me how even the most fundamental functions or statements in VB6 are not carried over to VB.NET. Below is an example of such fundamental code that runs fine under VB6 but will not produce the same results using VB.NET. While the code editor provides indents and line numbering that are helpful features how can VB6 programmers use the VB.NET version (find workarounds and fixes) when even the VB.NET converter and docs do not deal with such issues? 71.100.1.71 (talk) 19:25, 4 May 2010 (UTC)


 * VB6 and VB.NET are clearly quite different platforms; the VB.NET article discusses whether it should be considered a new version, or a different language altogether. This article covers how one might migrate from VB6 to .NET. The assumption that VB.NET is compatible (or even tries to be) seems to be false, as is a belief that the automated conversion is anything other a partial solution.  If one wants to develop for VB.NET, one needs to acquire the skillset necessary for coding for that platform. -- Finlay McWalter • Talk 20:37, 4 May 2010 (UTC)
 * If there is not a simple and straightforward fix for the code above, which will allow it to run under VB.NET then I'll start calling it Vista.NET and relegate it to the pile of software that sacrifices the real in the course of providing the illusion of having achieved the ideal. 71.100.1.71 (talk) 21:21, 4 May 2010 (UTC)


 * It's a totally different language that happens to preserve some of the basic grammar of VB but that's about it. It's not the same and whatever you want to call it, don't be confused by trying to consider it VB. Personally I think it is a half-language between VB and, I don't know, Java or C or something... superficially VB-like but with all of the pain-in-the-neck qualities of the latter that VB was supposed to let you avoid (like having to deal with memory management and complicated threading systems and etc.).
 * With that specific code... I think you will have to declare and type your variables explicitly, for one thing. I think everything else should handle reasonably similarly... that is, I think for/next loops and arrays handle the same way. You may have to explicitly include the Strings and Math modules in order to use Mid and Int. I can't recall; I haven't used VB.NET for awhile and it always takes a little work to get my brain back in gear on it. --Mr.98 (talk) 21:37, 4 May 2010 (UTC)
 * Don't worry about it. I've pretty much reached my conclusion that if my kids want to follow in my footsteps to do it somewhere besides where I work. (No offense Microsoft.) 71.100.1.71 (talk) —Preceding undated comment added 22:01, 4 May 2010 (UTC).
 * Interesting. I looked at that code and thought you could get away with the typing. I'm pretty sure .NET will baulk at that Mid function, which I'm not entirely sure actually exists in .NET. It acts as String.subString, right? So you need to use that (with an implicit conversion, by the looks of things). Just my 2c though. - Jarry1250 [Humorous? Discuss.] 21:48, 4 May 2010 (UTC)
 * Reading the docs such as they are I for awhile tried to write my own functions for mid and int and strdup, etc. but these all had the same coding problem. I tried Option Explicit and watched the warning messages carefully and fixed the problems they cited one by one. Yet VB.NET could still not run this simple code. Maybe Microsoft farmed out VB to be upgraded and improved and brought into the future by persons already in a different reality far from where the real future will actually go. 71.100.1.71 (talk) 22:06, 4 May 2010 (UTC)


 * VB.NET does have Mid and Int functions but I think you have the import the relevant library. I think you have to do something like this:
 * Imports VB = Microsoft.VisualBasic
 * And then refer to them as VB.Int and VB.Mid. Maybe. Again, I'm rusty (and on a Mac, so trying it before telling you is hard!). Anyway, again, I agree VB.NET is kind of stupid. At this point I would just tell people to learn C or Java; all of the simplicity of VB has been sucked out of the language, and if that's the case, what's the point. --Mr.98 (talk) 22:46, 4 May 2010 (UTC)
 * Perhaps you are thinking of importing Excel using the import directory rather than math. In any event I completely agree that replacing such basic (literally) statements and functions with some sort of system contrived to supposedly add functionality and instead making the language disfunctional is never the way to go, unless of course you are trying to form a C# or C++ Guild and want to destroy all competition. 71.100.1.71 (talk) 23:14, 4 May 2010 (UTC)
 * I'm not sure I understand your first sentence. Did you try adding the "Imports" line to your code and changing the functions as I suggested? That ought to work... --Mr.98 (talk) 01:05, 5 May 2010 (UTC)
 * Yes, but "import" was treated as an undeclared variables. 71.100.1.71 (talk) 10:58, 5 May 2010 (UTC)
 * I wouldn't really call it is dysfunctional language. :) VB.Net has a different set of priorities to VB6 - the main change being a requirement to run under the .NET framework, and that therefore entails that it needs to be Object Oriented and to employ stricter typing, amongst other alterations. These are significant changes, effectively making a new language with a common core syntax. I tend to see VB.Net as an option for VB6 developers: if you wish to migrate to the new OO and .Net paradigms, here is a language which retains the syntax, more-or-less, of what you used before, even though it's not the same language, and this may make the transition easier. However, from my perspective, the move to OO is actually the hardest part of the change, so keeping the syntax consistent offers little advantage. The problem with the above code isn't so much that the various functions don't exist in VB.Net - they do, (although they need to be called differently), and thus the algorithm could be implemented in VB.Net using very similar statements - but that the code is unstructured, and would need to be converted to OO. Thus I tend to argue that it makes as much sense just to move to C#, as if you need to learn a whole new paradigm anyway, you might as well pick up a C-type syntax while you're at it. :) - Bilby (talk) 23:40, 4 May 2010 (UTC)
 * I could find no reference as to how int might be called differently except perhaps as a function. Calling it as a user function did not change the results. I use to write short C routines so perhaps that is the better course. 71.100.1.71 (talk) 23:51, 4 May 2010 (UTC)
 * Easiest method to convert a numerical value to an int is just to explicitly cast it, along the lines of x = (int)k, although that can run into problems with certain data types. Alternatively, you can use the ToInt32 method on the Convert class: x = Convert.ToInt32(k). - Bilby (talk) 01:23, 5 May 2010 (UTC)


 * The above code is very confusing to humans — not just the VB converter. The very first line starts out with a loop from i to v ^ k, even though both v and k are not defined, yet. You should also indent your lines so we can see more clearly what code is inside what loop. Also place more line breaks between lines. You should define everything at the top of the program and place comments throughout your code. Use Dim to define variables and strict type them to a specific data type — for example: Dim i As Integer. Also give variables meaningful names. Instead of "k" you could use "intTotalSales," for example. VB.NET is much stricter than VB when it comes to defining things. I congratulate Microsoft for rejecting the above code. The only way you could ever get away with using such constructs is by turning both Option Explicit and Option Strict off. If you follow these guidelines, your code will be converted with fewer complaints by Visual Studio.--Best Dog Ever (talk) 01:51, 5 May 2010 (UTC)


 * My apologies. The above code was intentional reproduced here as incomplete code and does not show the declarations, comments and indentations you mention. The code was intended to produce a test pattern only. Remarkably, however by skipping the input routine and loading the array internally by assignment with filtered (trimmed) data the above code was finally able to work, the data apparently being the real culprit. Thus some hope has been restored that VB.NET might be a useful platform if it will run in a distributed or cloud environment. However, because the speed problem is now quite evident I am back to the original quest of finding a large and fast enough system (a distributed or cloud system) on which to run the VB.NET code. 71.100.1.71 (talk) 11:05, 5 May 2010 (UTC)

Tinkering with browser source code
Is any browser simple enough so that an amatuer like myself could tinker with the source code to rectify irritating features or quirks they dislike, or conversley implement special features of their own? Or is it all way too complicated? Can it also be done in ten minutes rather than ten hours? Thanks 92.29.62.241 (talk) 19:30, 4 May 2010 (UTC)
 * I don't know anything about programming, but perhaps developing a Firefox addon would be easier than delving into browser source code. The Mozilla website has guides for creating extensions that should help too 82.43.89.71 (talk) 19:35, 4 May 2010 (UTC)


 * A modern browser has a huge number of features, not all of which are necessary for the basics of browsing the web. It really helps to know a few things about what a "web browser" actually is: at its core, it has a network layer that understands the HTTP protocol, and a layout engine that understands how to convert HTML into a graphical "picture" with text and images in the correct locations. Modern browsers like Firefox also do much more - storing bookmarks, managing complex tabbed browsing, handling sessions, saving graphical preferences, enabling rich multimedia plugins, understanding other protocols (like HTTPS, FTP, and others), hosting a scripting environment for JavaScript and other languages, and so forth.
 * As such, although Firefox is totally open-source, it is horribly inaccessible if you are a novice. You might want to look at a bit more primitive "browser", such as the Jakarta HTTPClient. You should be familiar with programming and the basics of how network sockets work; but you can read this sourcecode and documentation without being overwhelmed by the "peripheral" features. Note that HTTPClient does not have a rendering / layout engine. Alternately, Gecko (layout engine), which drives Firefox, is a powerful HTML layout utility, but is not a "browser" because it has zero internal knowledge about network connections - it just parses HTML documents that you feed into it, and makes them look nice on a screen.  You can play with Gecko independently from the full Firefox source code.
 * Lastly, as has been pointed out above, if you just want to tweak up an irritating feature of Firefox, you might be better off developing a Firefox extension or learning to use Greasemonkey. These will give you a lot of control over user-interface without requiring in-depth knowledge of the system internals.  Nimur (talk) 19:44, 4 May 2010 (UTC)


 * I think this falls under the category of, "if you have to ask, the answer is probably 'no.'" I mean, I wouldn't know how to modify the source to be able to change specific features or add new ones, but I do know that even optimistically this would fall way outside the "ten hours" requirement (and certainly more than ten minutes)—even small, trivial changes are likely to take you a lot longer than that if you don't understand how the whole system works together. Even developing a simple Mozilla extension for Firefox is likely to take you more than ten hours the first time around. If you had the requisite knowledge to do this, you wouldn't have to ask us how hard it is... you'd know. --Mr.98 (talk) 22:40, 4 May 2010 (UTC)
 * What does fall closer to the 10 minutes mark is tweaking firefox to your liking using about:config, there are really quite a lot of options. From there you can move onto modifying the appearance of the browser by modifying the userchrome.css file (e.g. removing redundant context menu items you never use), then you could move onto javascript using greasemonkey or similar and finally onto full blown extensions (perhaps starting by converting your favourite greasemonkey script into an extention). Tweaking your browser to exactly how you like it is more of a labour of love than a 10 minute fix though. If you want a quick fix, quite often other people have come up with solutions and posted them online - try googling specific things you want to change or if that fails ask here (make sure you know roughly what the code online does though - people can use client-side javascript to spread malware). Firefox is probably the most customisable browser, so it's probably your best bet for this sort of thing. Here are a few links you might want to check out:


 * http://www.mozilla.org/unix/customizing.html
 * http://lifehacker.com/197715/customize-firefox-with-userchromecss
 * http://kb.mozillazine.org/Editing_configuration
 * http://kb.mozillazine.org/About:config
 * 131.111.185.69 (talk) 09:26, 5 May 2010 (UTC)

How do I determine the URL and then capture a live video stream? (using vlc).

 * I have WindowsXP and use the graphic interface of the (free) "VLC media player 1.0.5 Goldeneye" from videolan.org. I also use FireFox 3.6.3 with the installed add-on: "Video DownloadHelper 4.7.3"

Why: I need to save video streams because, in most cases, such content tend to be lost (moved, completely removed or no longer freely availabe) before the time I want to take a look at it again.

What: Now I want to capture and save some video streams from a public broadcaster (NRK) web-TV site. For instance the stream on the page: http://www.nrk.no/nett-tv/klipp/630814. (Which is a norwegian language version of a BBC documentary on Supervolcanoes). When browsing it in FireFox then the web page first, automatically, determines the speed of my internet connection and then, whithout any problems, shows me the video.

Problem: Video DownloadHelper seems to ignore that there is any video-stream on the page. And if I use VLC and choose: "Media -> Open Network Stream" and then enter:  "http://www.nrk.no/nett-tv/klipp/630814" in the address field, then no video is shown (or saved).

Questions:
 * 1) How do I get VLC to sniff out the correct URL of the video stream itself (as opposed to the URL of the web page where the video is embedded)?
 * 2) How do I get VLC to save the stream to my harddrive?

Excuse: I know, I know: "RTFM!", but I have looked at what I believe to be the most relevant help page: http://wiki.videolan.org/Documentation:Streaming_HowTo/Receive_and_Save_a_Stream and I am overwhelmed and do not understand anything! Partly because I do not know much at all about video streaming in general, and partly because unfortunately I do read very slowly, therefore figuring out this problem seems a somewhat insurmountable task to me. ;-(

Request: If someone would tell me exactly what I have to do &mdash; preferably by less than a 143 words ;-) &mdash; then it would be really helpful to me! Could you please help me? --Seren-dipper (talk) 21:19, 4 May 2010 (UTC)

82.43.89.71 (talk) 21:36, 4 May 2010 (UTC)


 * I can't read Norwegian, but does it say that it requires Microsoft Silverlight for the videos? If so, you might be out of luck. I don't think VLC can save Silverlight streams (which are encrypted under proprietary DRM) at the moment... Googling around, I'm not sure any of the many stream-saver programs can, yet. Maybe someone else knows better though. You could always use screen capture software to grab the video as it streamed (e.g. CamStudio), though this has its downsides (large files, takes a lot of active CPU time). --Mr.98 (talk) 22:34, 4 May 2010 (UTC)


 * I believe that SilverLigt is not the problem.The web-tv site (http://www1.nrk.no/nett-tv/) has a settings-page (Norwegian link name: "Innstillinger") where one may choose between two players:1.Silverlight. and 2.Windows Media Player.I had chosen Windows Media Player, but I am lost so I still need a helping hand with VLC media player to make it capture and save a stream... --Seren-dipper (talk) 03:09, 5 May 2010 (UTC)


 * I don't know whether this will help you but for what it's worth, I don't personally recommend Firefox or VLC for this purpose. Perhaps you can find something which makes it work, I admit I haven't tried in a while however last time I looked nothing seemed that great, some sites have a tendency to make it as difficult as possible and many tools don't appear to manage well with such sites. But when trying to download any tricky stream, I usually use Internet Explorer combined with Orbit Downloader. This works well for most flash video sites and also works whenever I've tried it with WMP and other such streams, which isn't often but I did just check it on the Norwegian site and it seemed to download fine (well I only downloaded the beginning).
 * Now if you do choose Orbit, there are two options. You can use the GetIt button but that tends to be annoying so I recommend instead don't enable that.
 * Instead use the Grab++ tool. You can use this by going into the 'tools' option of Orbit after installing it. Once you have started the Grab++ tool, go to the page you want to download the video from with Internet Explorer (Firefox sometimes works but sometimes doesn't) and start playing the file. If you are already at the page, you should refresh and remember to start playback. Then go back to get Grab++ tool, in this case you should see a WMV file (the file extension i.e. the thing at the end of the file should be .wmv) which is what you want. If you aren't sure, you can often go by size since the file you want would usually be over 1 MB probably significantly large whereas other files will usually be small.
 * Once you have the file, you can view it with VLC or whatever you want. You only need to start Orbit when you need it, and I would also recommend you disable or don't install the Firefox plugin since it has some annoyances. Also while you can use Orbit as a general download manager I've found it annoying in numerous ways so I don't recommend that.
 * 11:09, 6 May 2010 (UTC)

HDMI Audio Problems
I am attempting to connect my laptop to my tv through HDMI. The video works fine. The audio, however, is not working at all; although, it has worked in the past. I changed the default playback device to the digital output and nothing comes out even though audio is playing.

I am running win 7 Home Premium (64-bit) on an HP Pavilion dv6700 (p/n: KL090AV) with BIOS version F.59. My audio driver is Realtek HD Audio.

HP is not helpful and keeps telling me that since I upgraded to windows 7 from the factory installed Vista, they can't help me. (One was so bold as to recommend that I reinstall Vista!)

Thank you! --omnipotence407 (talk) 22:54, 4 May 2010 (UTC)


 * Dell did that to me once, too, where I upgraded the version of Windows and they started refusing to help me until I jumped up and down and insisted it was a hardware problem and demanded to talk to a supervisor and so forth. Super frustrating.  But this may be a path to getting them to help you.  On, perhaps, an external USB drive.  Install Vista on an external USB drive and install the HP-provided drivers, and if the problem is still occurring then ring up HP and raise a fuss.  Comet Tuttle (talk) 00:39, 5 May 2010 (UTC)


 * This is a good idea, always best to try to isolate the problem. Also, call back HP and "play dumb," pretending you are using Vista.  See what troubleshooting steps they take you through.  Having done front-end tech support for a number of years, I can tell you that the guy/gal on the other end's goal is to get you off the phone as soon as possible.  Mentioning the upgrade will cause them to automatically give up on troubleshooting steps.  Also, I know it's not idea, but you can probably still pipe audio out of your headphone port into your TV with a $4 cable. --rocketrye12talk/contribs 22:18, 6 May 2010 (UTC)