Wikipedia:Reference desk/Archives/Computing/2009 September 6

= September 6 =

SQL server problem
When I try to go to http://www.hscl.cr.nps.gov/insidenps/report.asp?STATE=&PARK=&STRUCTURE=&SORT=&RECORDNO=6534, a National Park Service webpage, I get an error message as follows:"Microsoft OLE DB Provider for ODBC Drivers error '80004005' [Microsoft][ODBC SQL Server Driver][TCP/IP Sockets]Specified SQL server not found. /insidenps/global.asa, line 13"Is my computer missing something, or is this a problem with the NPS website? I've accessed pages like this before, including this page just two days ago, and never had this problem before. Nyttend (talk) 01:27, 6 September 2009 (UTC)


 * Your computer isn't missing anything, theirs is having a problem. My guess is that they keep their database on a specialized server that has somehow crashed or been turned off. (I would wait until Tuesday to expect anyone to look at it again, given the three day weekend and all.) --98.217.14.211 (talk) 01:38, 6 September 2009 (UTC)


 * As its name implies, SQL Server is a server side product. Nothing is required on the client to run.  If you can find the webmaster's or tech support e-mail address, I would inform them of the error.  A Quest For Knowledge (talk) 01:48, 6 September 2009 (UTC)
 * Thanks for the explanations; as all had worked a few days ago, I was guessing that it was on their end, but I wasn't sure. I saw "Microsoft" as a possible indicator that my use of IE was the problem; for all I knew, they had changed settings to something that IE didn't like.  If you'd said that was a browser problem, I would have perhaps tried to find a Firefox-equipped computer.  Unfortunately, there's no way to find an address for a webmaster; most of the National Park Service website has been down today.  Nyttend (talk) 03:49, 6 September 2009 (UTC)
 * For example — I found that their report-an-error page was visible, but when I tried to submit a report, I received a "page-not-found" message. Nyttend (talk) 03:53, 6 September 2009 (UTC)


 * You shouldn't see any sort of explanation of the error at all. The web master should have disabled custom error messages in the web.config file. Imagine what a hacker could do by changing the URL and seeing the output from the server. Everything you see is just plain incompetence, and it's more common than you think.--S1kjreng (talk) 05:41, 6 September 2009 (UTC)


 * Honestly, just wait until Tuesday. I bet it will work again. Nobody is there right now and nobody is going to get to it before then. It's a government website, not a for-profit dot-com that cares about its uptime. They'll get to it. --98.217.14.211 (talk) 14:05, 6 September 2009 (UTC)

Intel E4300 vs. Intel T3400
I am about to buy a low-cost laptop. I have a choice between Intel E4300 and Intel T3400. Which is better? Intel has a comparison at its website. The E4300 has a faster FSB and larger L2 cache, but the T3400 is newer and has a lower "Max TDP". Which is better? Thanks! --Masatran —Preceding unsigned comment added by 59.92.36.231 (talk) 10:39, 6 September 2009 (UTC)
 * Are you sure you saw E4300 on a laptop? I thought the E-series is for desktop? --antilivedT 11:18, 6 September 2009 (UTC)

What redirects here?
How do I find out what articles redirect to another article? In this case I'm interested in Female ejaculation ?Rfwoolf (talk) 12:49, 6 September 2009 (UTC)


 * Use the "what links here" option - -- Finlay McWalter • Talk 13:59, 6 September 2009 (UTC)
 * Not not the What links here, cos how do I distinguish between which terms links there, and which redirect there?Rfwoolf (talk) 16:27, 6 September 2009 (UTC)
 * Ahh okay I see you can then filter out the links... Got it. Thanks Rfwoolf (talk) 16:28, 6 September 2009 (UTC)

Copying text from an Acrobat Reader PDF document
Acrobat Reader has something called their Text Select Tool. When I use it to select and then copy text, and then paste that text into, for instance, TextEdit, the results are strange. The results require a lot of cleaning up. Here is an example of how something comes out:


 * Consult the ser onsult servic vice and suppor e support inf t information that came with y ormation your i our iMac f ac for or
 * '''inf information about ho ormation how t w to contac tact A t Apple f pple for ser or servic vice.

''' I have an Apple computer.

What is causing this? Is there a way to overcome this? Bus stop (talk) 14:57, 6 September 2009 (UTC)


 * What's causing it is how PDFs store text sometimes. Instead of storing it as a sentence, it stores it as a bunch of glyphs (letters) with specific positioning information (letter "C" goes here, letter "o" goes there), and doesn't necessarily know how they fit together. So in your case you have things like "inf" on one line and "ormation" on another, because the PDF doesn't actually realize that those are part of the same word and are sitting on the same line. My understanding of it is that it varies depending on what tool was used to make the PDF. Some programs do better than others. There's no way to overcome it that I know of. It's a problem with the PDF, not the reader. (This is assuming, based on the content, that it is not an OCR problem.) --98.217.14.211 (talk) 15:00, 6 September 2009 (UTC)
 * The problem seems to be much less pronounced with PDFs produced in 2009. I should have mentioned, that the PDF that I was trying to copy and paste from, was made in 2006. Maybe the problem has been to an extent addressed since then. Thank you for your help. Bus stop (talk) 15:51, 6 September 2009 (UTC)


 * I'm told another issue is that in a PDF file, the program that created the file may write out the text "information about ho" and then stop what it's doing, skip lower on the page to write unrelated text, then skip back to where it was and resume with "w to contact...". If this is so, it sounds difficult for programmers to write a PDF reader to reliably copy and paste text that happens to be contiguous visually.  Tempshill (talk) 01:44, 7 September 2009 (UTC)


 * The easiest way, in that case, would be to dump the page to an image file (PNG, for example), and then run OCR on that. It might work or it might not (or, most likely and even worse than not working, it will work with an accuracy of 98% or so...). Jørgen (talk) 08:58, 7 September 2009 (UTC)


 * Believe it or not it is now working. I went to Adobe's website and downloaded an updated version of their Adobe Reader. I guess I should have tried that in the first place. Thank you all for your input. Bus stop (talk) 16:21, 7 September 2009 (UTC)

Microsoft Excel Mode function
Is anyone familiar with the Microsoft Excel 2007 function called "Mode"? It is supposed to return the mode in a set of data. Sometimes, however, a set of data has more than one mode. Nonetheless, Excel returns simply one (and only one) result. That is, Excel returns only one of the modes, but not all of the modes. Does anyone know which of the several modes it reports? Is it the highest numerical mode? The lowest? I can't seem to find any rhyme or reason in how Excel computes the mode in a situation where the data set has several modes. One of my thoughts is that it goes through each data point, one by one. As soon as it finds one mode (any mode), it simply reports that. And any subsequent additional modes that it finds, it simply ignores. But, that is just a guess on my part. Does anyone know? Thanks. (Joseph A. Spadaro, 6 September 2009)


 * Here is the official documentation for the function, but it does not explicitly answer your question. Maybe the best solution is to test on a few known data-sets to see if the behavior is consistent?  In this case, the best programming practice would be to assume that the value is "undefined" on such a data-set (or at least, that the selection of which mode is undefined), unless you can guarantee that it is always (e.g.) the lowest of the possible valid choices.  Nimur (talk) 21:20, 6 September 2009 (UTC)


 * Fiddling about in Excel 2003, I see there is the same problem. If there is more than one mode, it looks like Excel reports the first mode in order of the position of the first element of that mode in the array - not necessarily the first mode. For example:
 * mode(2,5,3,5,1,2,1) = 2
 * The "first mode" would be 5 but it reports the 2. Do you get the same result in Excel 2007? Zain Ebrahim (talk) 10:56, 7 September 2009 (UTC)


 * Thanks to both of you for the input. To Zain Ebrahim ... I think that your theory is correct.  Yes, I do have the same problem and the same results in Excel 2007 as you do in Excel 2003.  To test your theory ... I tried the following functions and got the following results (among several other trials):
 * mode(700,333,8,8,8,700,111,555,700) ... Result = 700 (as your theory predicted).
 * Using the same exact data set, in different order, yields:
 * mode(8,700,700,700,333,111,555,8,8) ... Result = 8 (also, as your theory predicted).


 * Now, this leads to my follow-up question (below). Thanks.   (Joseph A. Spadaro, 7 September 2009)

Follow up question:

I thought that the Mode function also works on text data (that is, words) ... in addition to numerical data. In fact, in the Remarks section of the official documentation, it states: "Arguments can either be numbers or names." Therefore, as a test, I tried the following data set:


 * egg, star, star, star, tree, orange, frog, egg, egg.

And I expected to get a mode of "egg" ... but I got an error message instead (specifically, #N/A). I tried formatting the above list of words as "general format" and also as "text format" ... both formats returned an error message for the mode function. What am I doing wrong? Thanks. (Joseph A. Spadaro, 7 September 2009)

I'll think you'll find that Names in this context refers to Named Ranges. 86.165.115.70 (talk) 15:50, 7 September 2009 (UTC)

MPEG-2 and AVCHD Video Files
Dear All,

Can anyone tell me a way to convert to either of these two file formats if you have a camcorder that doesn't use them? Mine will only record in a strange format for which the file extension is .mt2s

If someone could recommend a program that would convert (especially for free) I'd be very grateful.

90.218.48.56 (talk) 20:22, 6 September 2009 (UTC)


 * Avidemux and Handbrake both support .mt2s files —Preceding unsigned comment added by Avrillyria (talk • contribs) 20:42, 6 September 2009 (UTC)

Sites - copyright
What's the best way to tell if a site is legit? Specifically (popups!) Is it a no brainer that this site is circumventing copyright (I don't want an angry letter from my ISP...)83.100.250.79 (talk) 21:20, 6 September 2009 (UTC)
 * When you say "legit", I think you mean whether accessing a particular website violates any laws. Since we don't know where you are or what you're doing, it's impossible to know; and we can't give you any legal advice (see our legal disclaimer at the top of the page).  If you're interested in learning about digital copyright law in general, we have numerous articles you can read.  In the United States, the Digital Millennium Copyright Act is the applicable legal standard; typically, Online Copyright Infringement Liability Limitation Act explains the liability of one party for the infringing acts of a second party.  Again, we can't interpret whether these apply to your particular scenario; or make any guarantee about the accuracy of the articles at any given time.  Nimur (talk) 21:37, 6 September 2009 (UTC)
 * obviously my intended activity would include downloading stuff via this site, what I'm asking is suppose I want to download a Rhianna (or Shakira song) (as examples), and I don't entirely trust that everyone using the site has read and digested the DMCA etc - so there might be some non 'legit' stuff on there.. How can I find out if a music label has released a song for free (as does happen - for publicity reasons etc I suppose) - is there a simple way I can find out the copyright status of a song.?
 * 83.100.250.79 (talk) 22:14, 6 September 2009 (UTC)


 * No, there is no simple way. Everything must be assumed to be copyrighted unless you know otherwise fairly explicitly. Anyway, that site is almost certainly violating copyrights. Whether they are breaking laws depends on their jurisdiction. (You can tell this pretty easily by reading its disclaimer, which is a standard "we're not violating copyright, our users are! nothin' we can do about that!" that all sites who habitually violate copyright have posted. In this case I'm not sure it is even true; it sure looks like the content is hosted on their server, which makes them liable.)
 * Now the upside for you, as a user, is that the only people who know, at the moment, that you've downloaded anything is the skeezy site. It's not the same thing as the P2P software where one of your "peers" might actually be a copyright watchdog. So the odds of getting a letter from your ISP are pretty low. But I am not a lawyer, yo. --98.217.14.211 (talk) 23:53, 6 September 2009 (UTC)


 * There is an easy way - if the site says something like "this song is provided copyright-free" then it is - if not, not. Everything that anyone writes, draws, paints, composes, sings, publishes, sculpts, etc is automatically copyrighted.  There is no need to put a little 'c' in a circle or register it someplace.  So unless there is a specific statement (such as at the bottom of every Wikipedia page) that says that there are some special terms and conditions, then it's safe to assume that the work is indeed copyrighted and that you need permission - or some kind of license - in order to be able to copy it.  Even if you ARE allowed to copy it - you may be required to accept some additional terms and conditions (as is indeed the case with the Wikipedia GFDL and CC-BY-SA 3.0 license terms) - which may prevent you from giving copies to your friends or putting the thing on your website or whatever - even though you were allowed to download it for free. SteveBaker (talk) 15:56, 7 September 2009 (UTC)