Wikipedia:Reference desk/Archives/Computing/2008 June 5

From Wikipedia, the free encyclopedia

Computing desk
< June 4 << May | June | Jul >> June 6 >
Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


Contents


[edit] June 5

[edit] Windows Vista or DOS?

Hello. I was considering replacing my operating system. I have narrowed it down to two choices: Windows Vista Ultimate and MS-DOS, version 1.0. Which is the better choice? Thanks.--O4irtj (talk) 01:29, 5 June 2008 (UTC)


Go with Linux, sir. You can't go wrong! —Preceding unsigned comment added by Duomillia (talk • contribs) 01:30, 5 June 2008 (UTC)
What is Linux?--O4irtj (talk) 01:33, 5 June 2008 (UTC)
Here you go, it's the best of both worlds! --antilivedT | C | G 06:14, 5 June 2008 (UTC)
To answer the original question, which is best depends on hardware and requirements. -- Q Chris (talk) 07:22, 5 June 2008 (UTC)
I think it's safe to assume that the original post is a joke of some sort, but I'm not sure what the point is. If it is that Vista has a bucketload of issues, then yeah, we know. But I'd bet so did MS-DOS in its first version, and it is clearly inadequate for most forms of modern computing. Even if the OP insists on a minimalistic, no-nonsense platform, I'm sure there are much better possibilities (MS-DOS 6.22 comes to mind, as well as Unix-like command-line builds). Ultimately, the joke is on the OP, since he allegedly "narrowed it down to two choices" without really being aware of the possibilities (e.g. Linux, the BSDs) and ignoring those he probably knows (Windows XP, Mac OS). -- Meni Rosenfeld (talk) 08:08, 5 June 2008 (UTC)

Thanks for the replies. That "Linux" Duomillia mentioned looked good, so I decided on the first version that I could find: Slackware. So, I installed it and it's working like a charm! I ported all of my Windows applications over to it. I never knew that I could be so productive. Thanks again guys.--O4irtj (talk) 08:30, 5 June 2008 (UTC)

I was considering this (DOS or XP) on an old computer with no working hard drive, but a working floppy, DOS started looking attractive. Is there a linux that can boot off a 1.44 meg floppy on a PC? Graeme Bartlett (talk) 12:13, 5 June 2008 (UTC)
Coyote Linux. BTW- I have MS-DOS 3.3 on a 5¼ floppy. --—— Gadget850 (Ed) talk - 12:20, 5 June 2008 (UTC)
The OP found a hyperdrive across the learning curve and got productive in using Slackware Linux (GNU/Linux) in seven hours and a minute [my bad, cut that down to less than six hours and fifty seven minutes] from not even knowing what it meant. I want to recommend that the OP for the Guinness Book of World Records. (jk) Kushal (talk) 13:42, 5 June 2008 (UTC)
Sniff, sniff, I smell troll. Sandman30s (talk) 14:00, 5 June 2008 (UTC)
At best he's looking back through rose-tinted glasses: DOS version 1 is dire. Forget all the nice features you get in cmd.exe - command history, concurrent pipes, even tab completion. Hardware support is through the impossibly difficult config.sys; sound cards are problematic and hard drives use FAT12. Security and networking are non-existent. --h2g2bob (talk) 22:25, 5 June 2008 (UTC)
Hard drives are FAT16; floppy drives are FAT12 (though the user's mentioned MS-DOS 1 didn't have hard disk/FAT16 support). You can get FAT32 in DOS as well, if you download FreeDOS, illegally download the pulled DR-DOS versions (8.x), or get the extracted MS-DOS from Windows 9x. JeremyMcCracken (talk) (contribs) 02:15, 12 June 2008 (UTC)
??? MS-DOS is a disk operating system, not a hardware abstraction layer. Why would it provide sound support? That's the job of applications that need sound. Config.sys and autoexec.bat aren't too hard to work with, unless you're trying to fit a great many TSRs in while still leaving enough conventional memory for large applications. Security is only a concern if you're running a multi-user system or if you're running server applications, neither of which you're likely to be doing on MS-DOS 1. I admit the lack of support for FAT-16 and network drives is a problem, as it also makes CD-ROM support impossible.
MS-DOS was designed around the philosophy of "get out of the way and let the application do what it wants", as opposed to the managed environments of modern operating systems. --Carnildo (talk) 20:14, 6 June 2008 (UTC)

Between those two choices... DOS. Don't get MS-DOS 1 though, get DR-DOS 7.03; it's available online. Yes, I'm serious, I hate Vista JeremyMcCracken (talk) (contribs) 02:15, 12 June 2008 (UTC)

[edit] Way to track info of person viewing the web page

Is there any way to track the information (location or ip address ) of the person visiting my homepage ? If yes, how? I have learned that there is a way to keep track of number of persons visiting the homepage.But don't no if the other thing is possible.... —Preceding unsigned comment added by 137.132.250.10 (talk) 04:30, 5 June 2008 (UTC)

Yes, there are ways of tracking the ip/location of visitors. I can't tell you how exactly, but people have already written code that does just that. A simple google search gave this, this and this. Leeboyge (talk) 07:55, 5 June 2008 (UTC)
If you want really nice and pretty statistics, Google Analytics is a lot of fun and easy to use. It can draw a whole map of the world and show you how many visitors come from what country, what pages they look at, how they find your page in the first place, etc. For my own use I installed BBClone on my server (need to have PHP on the server), which is a simple little way to look at individual users (shows you how they get to the page, where they go on the site, etc.). It's harder to use and set up than Google Analytics, though. --98.217.8.46 (talk) 14:06, 5 June 2008 (UTC)
If you are running your own server, yes: look at your server logs. If your page is hosted by someone else, then it entirely depends what facilities your host makes available to you. --ColinFine (talk) 21:26, 7 June 2008 (UTC)

[edit] Google

I dislike Google's new feature: when you enter a phrase with quotes and if there are no results, it just shows the quoteless results. How can I turn this feature off? Interactive Fiction Expert/Talk to me 23:20, 3 May 2008 (UTC)

Your observation that "it just shows the quoteless results" is not accurate. It displays a warning icon and a message indicating that the search returned no results. The unquoted results are displayed below the warning message. I do not know of any way to turn it off. -- kainaw™ 00:58, 4 May 2008 (UTC)
I don't think you can turn it off. Google says it did not find the results you wanted and tried its best to come up with the results that seemed most relevant. Kushal (talk) 15:29, 4 May 2008 (UTC)
A possible workaround is to append something like -madeupwordthatgetsnohits1234dfsdsdtgdc to your query; it shouldn't affect the results (assuming the word you pick doesn't get any hits, which this one of course will as soon as Google next indexes this page), but it confuses Google enough that it won't automatically remove the quotes even if there are no hits. —Ilmari Karonen (talk) 04:20, 5 May 2008 (UTC)
madeupwordthatgetsnohits1234dfsdsdtgdc does not give any results ... so far. Kushal (talk) 03:56, 6 May 2008 (UTC)
However since the reference desk is indexed it does now Nil Einne (talk) 13:54, 8 May 2008 (UTC)
You could also try adding a trivial word like a or the to get much the same effect, so long as your target pages contain any reasonable English text. A silly but more reliable workaround is to use the OR operator: "foo bar" OR "foo bar". --Tardis (talk) 20:10, 5 May 2008 (UTC)

Please just tell me how to turn this feature off. The -madeupword strategy is silly - how am I supposed to know that there will be no results until after my first try? And the "this" or "this" one - it don't work neither. Please help me turn the frigging feature off - I don't need "help" finding results with those words "scattered" across the page. Interactive Fiction Expert/Talk to me 11:17, 5 June 2008 (UTC)

What makes you believe that the Google developers answer questions here? This is a reference desk. The answer to your question is not in Google's online instructions. Therefore, nobody here can provide a reference to the answer. The answer you seek is at Google. You have two sane choices: Stop using Google or call Google and try to find someone there to give you an answer. You have many insane choices, such as demanding that people who don't know the answer give it to you. -- kainaw™ 12:03, 5 June 2008 (UTC)
Try the Google Web Search forum at http://groups.google.com/group/Google_Web_Search_Help --—— Gadget850 (Ed) talk - 12:09, 5 June 2008 (UTC)
How DARE you personally attack me!?
  • "What makes you believe that the Google developers answer questions here?" I believe nothing of the sort.
  • "This is a reference desk." Yes, and just about anything can be asked here, except maybe questions whose answers no RD users know. As far as I know, this is no such question.
  • "Therefore, nobody here can provide a reference to the answer." As above, maybe one person can.
  • "You have two sane choices: Stop using Google or call Google and try to find someone there to give you an answer." 1. Google is the most recognisable search engine. No way are you stopping me from using it. 2. What is Google's phone number? Besides, since I'm from Australia, it'll probably cost heaps to call Google. And I'm 15 years old, for God's sake. I'm not the type who is supposed to talk to a bunch of high adults.
Unless you do not want to use a proprietary protocol that probably has security holes the size of football fields, you can use Skype to call Google's 1-800 number. (Maybe Goog411 will help you find Google's number.) However, I don't think you can get an answer to your question on the phone. Kushal (talk) 13:36, 5 June 2008 (UTC)
"High adults"? Is the implication here that Google employees are all stoners that answer their phones high? This is an amusing, but unlikely image. APL (talk) 15:50, 5 June 2008 (UTC)
  • "You have many insane choices, such as demanding that people who don't know the answer give it to you." As above, I'm not addressing anyone in particular, just hoping that there is someone here who does know the answer.
Interactive Fiction Expert/Talk to me 12:35, 5 June 2008 (UTC)
Both of you— please stop. Either someone will come up with an answer or the question will go unanswered. --—— Gadget850 (Ed) talk - 12:38, 5 June 2008 (UTC)
Out of curiosity, why is this feature a problem anyway? As it tells you it's done the quoteless search, what possible harm does it do? ~ mazca talk 14:07, 5 June 2008 (UTC)
I hate to admit to agreeing with IFE, but this 'feature' bugs me too. I often don't notice the warning. (Who reads the text before the results?) Since I sometimes google for something just to see if it exists, this is very slightly annoying. Nothing to get angry about though. APL (talk) 15:50, 5 June 2008 (UTC)
You'll need to substantiate your claim that my suggestion (the "this" OR "this" one) doesn't work, because I have evidence that it does (as of this writing). Perhaps you didn't know that (unlike search terms) Google's "OR" operator is case-sensitive? I gave it in the correct case in my suggestion. For that matter, my "trivial word" suggestion also seems to work. Please realize that I would not have suggested these things if I hadn't already tested them myself; I wouldn't need to speculate about the behavior of a publicly available service. I also fixed your quote of the previous post to be legible. --Tardis (talk) 15:19, 5 June 2008 (UTC)
This is a handy trick. I notice that it also works when the first string is a null string. (Compare "" OR "The anyone Encyclopedia" "" OR "The Free Encyclopedia" "The Anyone Encyclopedia" ) I may have to look into modifying my Firefox search box so that it always does that when I search for something quoted. APL (talk) 15:50, 5 June 2008 (UTC)
You don't even need the quotes: OR "The anyone encyclopedia". Algebraist 15:57, 5 June 2008 (UTC)
From Google's suggestion that you use "OR" as an operator when you send that search, I surmise that you're actually using my other trick of including a trivial word. I realize now that Google actually entirely ignores those words; you can use "+or" to require such a tiny word, but it's hard to even find an example where that changes anything. But that's good: the trivial-word trick will thus never hurt you (by missing a page that happens to lack the addition). --Tardis (talk) 17:07, 5 June 2008 (UTC)
It seems like you're absolutely right. I just edited line 43 of the XML file for Firefox's google widget to say <Param name="q" value="the+{searchTerms}"/> instead of <Param name="q" value="{searchTerms}"/>. Now this won't ever bug me again. If IF Expert uses Firefox I recommend that he do the same. Thank you, Tardis. APL (talk) 19:59, 5 June 2008 (UTC)
Unfortunately, I don't use Firefox. I use Internet Explorer. Interactive Fiction Expert/Talk to me 09:55, 6 June 2008 (UTC)
Doing this in IE is really easy. Go to the dropdown menu next to the search box and choose 'find more providers'. Paste http://www.google.com/search?hl=en&q=TEST+OR+TEST (or whatever method you think works best; I haven't experimented much with those mentioned above) into the url field, give it a name, and click on install. Algebraist 07:45, 8 June 2008 (UTC)
As an update adding the word The to the begining of the search does seem to effect the results. (king and The King ) The odd thing is the estimated number of results. I have a hard time imagining that there are 474 Million pages that contain the word "king" but not the word "the". APL (talk) 18:34, 9 June 2008 (UTC)

[edit] Join & split videos

I need a good free or open-source application to join and split video files, mostly AVI format. At work, I use Adobe Premiere on occasion, but this is a home project and I don't need anything that heavyweight. --—— Gadget850 (Ed) talk - 11:50, 5 June 2008 (UTC)

You might try VirtualDub, but it's a bit weak for any sort of editing. APL (talk) 13:43, 5 June 2008 (UTC)
It seems that I cannot import more than one video at a time in VirtualDub. Any ideas? Kushal (talk) 19:03, 6 June 2008 (UTC)
I've honestly found Quicktime Pro to be the best tool for quicky little video editing like splicing together files, removing bits, etc. It's not free or open source but it's cheap. There are number of open source NLEs but I've never been able to get any of them to work at all to the degree that would be useful for me. --98.217.8.46 (talk) 00:51, 8 June 2008 (UTC)

[edit] Threading on talk pages

Why doesn't MediaWiki have any better system to auto-indent discussions on talk pages? It's easy enough to prefix a paragraph with a colon, but counting out four or five colons in threaded discussions gets tedious. You can outdent, but you might not want to, and if you do you should probably indicate such to show that your comment is meant to be nested under another one. Am I missing something? It seems like a button could be added to the toolbar or some such to indent your comment under the one above. Fletcher (talk) 14:26, 5 June 2008 (UTC)

I think questions like that are better suited for WP:Village Pump, but when I am writing an extremely indented reply, I just highlight the colons from the prior section and copy/paste them to the start of my reply, then add a single colon. --LarryMac | Talk 15:15, 5 June 2008 (UTC)
The Wiki format is not primarily intended as a "discussion forum"; here on Wikipedia we use it extensively as one, with talk pages, ref desks, etc. Unfortunately this secondary use has some usability issues (as compared to, say, PHPBB, which is far less versatile but makes for cleaner messageboarding). Perhaps you could discuss or contribute to the MediaWiki development platform? Nimur (talk) 15:45, 5 June 2008 (UTC)
A threaded system is probably the best solution, see mw:Extension:LiquidThreads, although I believe there's opposition to it. x42bn6 Talk Mess 20:32, 5 June 2008 (UTC)
Best solution... for what? "If ain't (that) broke, don't (try to) fix it." The beauty of the Wiki system is that we can fix someone else's indents if we did so want to, or do something totally different, or whatever. Structured systems (including hardcoded threading) will just lead to inflexibility. (And do we really need just another Javascript button on the already cluttered and frankly useless toolbar? It would be simple to write a function that would just add a hard return and indent one more than the previous, but how many people would think to even look for it, much less use it?) To me the idea of adding hardcoded threading sounds like a programmer's idea of what is "best", not necessarily a user's solution. --98.217.8.46 (talk) 22:58, 5 June 2008 (UTC)
I have to agree with 98 on this one. Kushal (talk) 19:12, 6 June 2008 (UTC)
See the universal problem solving flowchart (BTW, as an aside, where did that thing first come from, and does it deserve an article?). --Prestidigitator (talk) 19:51, 6 June 2008 (UTC)

← No, you can't edit others posts, it's a breach of policy. CWii(Talk|Contribs) 02:40, 12 June 2008 (UTC)

[edit] Can't map network drive

When I try to map my who.hasfiles account to a network drive (in Windows XP Pro, SP2), I get this error:

The drive could not be mapped because no network was found.

What should I do? By the way, it worked just fine before I reinstalled Windows. (Reason for reinstall was corrupted %windir%\system32\config\system file, if you're interested.) Do I need to install an update or something? --grawity 17:51, 5 June 2008 (UTC)

Addition: While trying to add the folder as a "network place", I get:

The folder you entered does not appear to be valid.  Please choose another.

The "official" mapper tool says:

Cannot map the storage. Wrong name and password, or network problems.

No proxy set in Internet Explorer (7). (Which can access the site just fine, by the way.) --grawity 18:00, 5 June 2008 (UTC)

Addition: Windows eXPee inside a VirtualPC does exactly the same. I even tried disabling the firewall (KIS 7), nothing changed. HALP! --grawity 18:15, 5 June 2008 (UTC)

Okay, so it sounds like after you reinstalled Windows, your network drive doesn't work anymore. Does this sound about right?
Also, are you saying that from Internet Explorer, you can access the computer that the network drive is on?
Finally, can you ping the other computer? (Start->Run->cmd->ping the-other-computer's-name) Indeterminate (talk) 22:49, 5 June 2008 (UTC)
Sort of. The "network drive" is my who.hasfiles.com account, [1], which I try to map using WebDAV. It's not on the LAN.
I can open the account ([2]) with Firefox 3 (rc2), Opera 9.5, Internet Exploder 7, telnet and netcat. And I can ping it. But I can't mount it as a network drive in Windows Explorer.
In case you didn't understand it yet, I try to connect over the Internet using WebDAV, not over LAN. You can sign up at who.hasfiles and try it yourself. (Free account = 100 MB space.)
By the way, I installed XP SP3 -> no change.
--grawity 10:59, 6 June 2008 (UTC)
Addition: I can't even map http://127.0.0.1 (nc -vvlp 80 shows no signs of activity). I can connect to \\think\* (think is my computer's name), but that's all.
Addition: D'oh. net start webclient. Add the "resolved" template. --grawity 11:14, 6 June 2008 (UTC)

[edit] Booting an Intel MacBook from a USB drive

Is it possible to boot an Intel MacBook from a USB drive if the drive is made bootable? If so, how does one doing so without first booting into OS X? --213.140.21.227 (talk) 23:08, 5 June 2008 (UTC)

It is not an important question at the moment. However, I do have some questions. How would you make the flash drive bootable? and so on. Once these issues have been resolved, we can work out the details (maybe press option when booting?). I wish I could be of more help. I will try to do some googling and let you know. boot from USB drive on an Intel mac Kushal (talk) 19:17, 6 June 2008 (UTC)
Ted Landau posted a note at MacFixIt about booting Leopard from a USB flash drive [3], but it looks like it's no longer available to nonsubscribers. If I remember right, he found that it'll boot normally from a system installed on a flash drive provided the drive is at least 8GB; smaller than that, it refuses to recognize the drive as bootable. Also, I seem to remember that the first generation of MacBooks didn't support booting from USB (although this might've been changed by subsequent firmware updates).
As for how to boot from the drive: hold the Option key as you power the MacBook on; this this will select the firmware's Startup Manager [4][5], which gives you a list of bootable volumes to select from. Speaker to Lampposts (talk) 06:24, 8 June 2008 (UTC)

[edit] A script that searches out and responds to certain elements in a website?

Hi all. I'm only medium in the field of programming, but I was wondering if anyone could give me some resources, or some terminology to search for in, say, google, about a script that would:

  • Go to a website
  • Search out (in its source code, preferably), certain sections of text, ie words or phrases.
  • Depending on what they were, would then act accordingly.

Just for an example, say a webpage contained the word "fuck", then the script could create a frame that says "This website uses profanity." Or if a website is an e-book, with a certain page number, it could search through the text until it finds, say, "Page: 34" and then, in a frame, print out the links for page 33 and page 35 accordingly. Just examples, so you can figure out what I'm trying to ask here.

Again, any ready-made scripts (the simpler to understand and experiment with, the better) would be best, but also any resources towards learning how to do so, or terminology for such scripts so that I can search out and learn at leisure (my current searches are failing horribly) would be great!

Also, please do not direct me to Greasemonkey -- that is a third-party software, and requires firefox. I was thinking there could be something internet-based (ie, php, javascript or whatever), or something downloadable that would do it, not an extension of a piece of software that requires a certain internet browser.

Much help appreciated ! -=- Xhin -=- (talk) 23:42, 5 June 2008 (UTC)

Well, I'm not quite sure what you want, but my best guess is that you mean you want to intercept all outgoing web connections and add stuff to each page, depending on the content in it. Personally, I would probably use a proxy like squid (software) with perl to manipulate the text. For an idea of what I'm talking about, take a look at this [6]. If you're using Windows, you might be able to find caching web proxy software that will let you manipulate the html, but I don't know of any off the top of my head. Indeterminate (talk) 00:55, 6 June 2008 (UTC)
There has been a lot of work on querying web pages, such as WebSQL and Twig Queries. The goal is to create a simple syntax for querying content in web pages and performing proper functions, such as making a list of items or creating a new web page of results. In my opinion, the current work simply stinks. I was forced (in college) to study and implement many forms of this web querying thing. What I found was that all the existing methods were developed by math majors, not computer majors. So, while mathematically sound, they were not designed with a working knowledge of computers. In the end, they look pretty on paper but aren't worth the effort to implement. I'm considering describing a better standard for a thesis - a simple mix of standard SQL, RegEx, and HTML/XML. Since all three are already standard, users won't have to learn something weird. If you know all three, it wouldn't be hard to use. For example, if you know SQL and RegEx, you can already figure out what the following "web query" does: "select dob, count(*) as cnt from google:"john lennon birthday" where dob=/[0-9]{1,2}\/[0-9]{1,2}\/[0-9]{4}/ order by cnt desc limit 3" -- kainaw™ 02:49, 6 June 2008 (UTC)
From what I understand of what you are asking, you want to get the content for a web page, parse it, and act on any key words found. If you want to get down to the lower level stuff and do it yourself, you will want to look up sockets and the HTTP protocol. This will allow you to request a web page and receive its contents. Most languages have support for sockets, PERL being one of the easiest to use. If you don't want to do this yourself, you could probably find a package that would do the request for you. Leeboyge (talk) 05:19, 7 June 2008 (UTC)