Wikipedia:Reference desk/Archives/Computing/2011 January 2

= January 2 =

Subtitles in download
I downloaded a foreign film movie, which said that subtitles were included. I burned it to DVD I then tried to watch it. The image and sound is beautiful but no subtitles. So I took a look at the download folder and it has, separate from the avi file containing the movie, a file called "English.srt", which I opened with text editor and can see that it contains the subtitles and where they go, in a form like this " ". I've just now figured out that the .srt file extension refers to SubRip. Still, I haven't a clue how to integrate into the downloaded file the subtitles. Can anyone give me a step-by-step explanation of how to add the subtitles to the film? I am using an imac running snow leopard. I use toast titanium for burning.--141.155.143.65 (talk) 03:29, 2 January 2011 (UTC)

With VLC, I think you should be able to select the file under right click, Video-Subtitles track. THen choose it or select the file. I think it will work... General Rommel (talk) 06:17, 2 January 2011 (UTC)


 * http://tovid.wikia.com/wiki/Making_a_basic_DVD#Converting_the_videos (there's a GUI if you want it, and I assume you can specify a subtitle file from it, but haven't personally used it) ¦ Reisio (talk) 16:17, 2 January 2011 (UTC)
 * Sorry. You're speaking a foreign language. I explored the link. Have no idea what that means or how I could use it to add it to the avi file for burning. I was able to figure out what you meant about "VLC" (you really should link thinks like that, for the uninitiated) but that seems to me a program that might allow me to watch the file with the subtitles, maybe, but not to add it to the avi file for burning. Even if the former, once again, you have to assume your audience does not have your sophistication, which I do not.--141.155.143.65 (talk) 18:35, 2 January 2011 (UTC)


 * In that case, you might try DVD Flick; a tutorial including adding subtitles (item 11) can be found here: http://beginwithsoftware.com/videoguides/dvd-flick-guide.html (if this seems out of date, check these:  ) ¦ Reisio (talk)


 * These instructions seem to tell how it could be done pretty painlessly on a Mac as long as you don't care if the subtitles are "switchable" or not (most commercial DVDs have "switchable" subtitles that can be turned on or off; that method would burn the text onto the image permanently). It relies on using the Perian plugin. --Mr.98 (talk) 20:51, 2 January 2011 (UTC)


 * Heh, Mac. http://www.google.com/search?q=srt%20site%3Avso-software.fr ¦ Reisio (talk) 21:53, 3 January 2011 (UTC)


 * The OP said they are using an iMac. I tried to give them an example using the software they already had, on the computer they were already using. Does linking to a forum about for-pay software they do not own, for a system they do not run, help at all? I'm not seeing how. --Mr.98 (talk) 15:38, 4 January 2011 (UTC)


 * Yes I missed that and found it humorous. Mac users like paying for things. ¦ Reisio (talk) 18:11, 4 January 2011 (UTC)

Firefox 3 and the cache: can i force websites into it?
I have Firefox 3.6.13. Going to Tools > Options > Network(Tab) provides me with cache options including a window that offers the websites which are currently storing data in my cache. This window is empty. However, i am running on a slow connection and i wish to have a website in my cache as i visit it. The items i might need caching are mainly pictures and flash content (.swf)...

Is there a way to force/tell my browser to collect this data and store it in cache for later use so i don't have to wait all the time for things to load that are on almost every page of that site, etc?

63.26.250.109 (talk) 04:07, 2 January 2011 (UTC)


 * You could try setting the amount of MB the cache can use higher. Or try using a standalone web cache which will give you much more control over which sites are cached and how 82.44.55.25 (talk) 10:38, 2 January 2011 (UTC)

"Found new hardware  ACPI Uniprocessor PC"
After restarting my computer this morning I had a yellow bubble pop up with the above message. Soon afterwards I got other messages saying that the hardware had been installed and I needed to restart my computer. Did so, no problems. I use WinXP Sp3.

Perhaps it was related to last evening when I restarted my computer, which does not power-down the USB pendrives I have pernamently attached to my computer, and after that restart (yesterday evening) I saw what I think is a USB symbol on the bottom right of the screen next to the clock.

Can anyone explain what the above was all about? Has for example windows been installing a better USB driver or will it just be the same one as previously? Thanks 92.24.185.225 (talk) 11:26, 2 January 2011 (UTC)

meta tag problems with <meta name="keywords"...
Hi guys, I've posted on here a few times recently (lots of very helpful responses ;) ).

I've been adding meta tags for 'description' & 'keywords' to the pages on a site im working on. The description tag is working fine, but when I validate the page (Using the validation tool in Macromedia Dreamweaer) the keywords tag creates the following error message:


 * "the tag: "meta" doesn't have an attribute: "keywords" in current active versions.[XHTML 1.0 transitional]"

The keyword section code (with some of the keyword content cut out save some space):


 * > ...tower hamlets news" />

I've looked at the keyword code on other sites, and the code I'm using appears to be correct. When I begin typing "<meta name="....." one of the Dreamweaver auto-suggest options is "keywords". link to one of the live pages with the problem

Does anybody know what I'm doing wrong? Cheers Darigan (talk) 13:06, 2 January 2011 (UTC)


 * No form of the string 'keywords' occurs in the page you linked. Also they're almost completely useless, I wouldn't even bother adding them.  Read: Meta element ¦ Reisio (talk) 16:19, 2 January 2011 (UTC)
 * Hi Reisio - I deleted the Keywords & Descriptions a short while ago - They were screwing up the Dreamweaver interface (I was about to update this post before I got distracted by another query on this page). I've just had read of the page you linked to, I had no idea that the keyword data was so insignificant as regards SEO. If I may - Can I ask your opinion on Description metadata - Im aware that this shows up on Google SERPs, but does it have any impact on SEO? I always thought that both Keywords & Descriptions had good SEO effects if the onpage content reflected the words and terms in the Keywords & Descriptions. Thanks for your reply Darigan (talk) 16:41, 2 January 2011 (UTC)
 * See Meta_tag.
 * My suspicion was that the META tag wasn't valid with your DOCTYPE, but I haven't been able to find anything that suggests such is the case. Have you tried validating it with something other than Dreamweaver? That will at least tell if you if it is just Dreamweaver or not, or a symptom of some other typo. --Mr.98 (talk) 19:01, 2 January 2011 (UTC)
 * Mr.98 - Have you got a link for a tool that I might be able to validate it with (I trust Macromedia Dreamweaver to be fairly reliable, but am not convinced by these particular validator results)? (Cheers for the reply) Darigan (talk) 21:22, 2 January 2011 (UTC)
 * Try this one. --Mr.98 (talk) 23:15, 2 January 2011 (UTC)


 * It's hard to say what the problem is without seeing the code for the page. Could you paste the first dozen or so lines of code for the page?--Best Dog Ever (talk) 00:49, 4 January 2011 (UTC)

Google and Youtube serious security flaw?
I'm trying to figure out if this is in fact a seucurity flaw. Google and Youtube were separate and they now have a unified login where your Youtube account is linked to a Google account. Older Youtube users that have not been linked to a Google account (or perhaps have not been logged into since that change) can still log in as before with the Youtube username and password. Google account holders can also log in by typing there Google username (the part before @gmail.com only) and their password. This creates a situation where the exact same "username" can be entered, and the account logged into will depend on the password. This is no problem as long as the users have different passwords. The problem is, if the passwords are the same by chance, the first person to attempt to log in will "discover" the password of the other user on the other service. While this may seem as unlikely as trying to guess someone's password in the first place, I think this might be much more vulnerable to happening, and it can happen without any "attempt" to hack into the other account. If your kind of lost on all of this, I will give a quick scenario. John Smith has signed up for the username jsmith on Youtube a long time ago and his password wanting his password to be easy, chose asdfasdf as the password. After Google decided to merge the logins of Youtube and Google, a Google user, Jason Smith, who had by chance originaly signed up for Google with the username jsmith and password of asdfasdf decides to start using Youtube, learns that he can use his existing Google account, and enters the username jsmith and password asdfasdf. Bang. Problem. He just discovered the other user's name and password and can take over both accounts and change the passwords. I have myself, already carried this out as an experiment. The result is you are shown a page telling you to link YOUR two accounts. It is assumed that it's the same person. Even though both accounts when created, where on completely seperate websites, completely seperate services, and login processes. This is not that unlikely to happen! Especially vulnerable are people who have simple usernames, and simple passwords. Simple password is kind of your fault but simple username is not. Either way, you should not be exposed to this vulnerability. Does anybody else, who has read all of this, not see this as a security flaw? -- Roberto75780 (talk) 14:00, 2 January 2011 (UTC)


 * The same thing happened on Wikipedia with Help:Unified login. An account on another Wiki with the same name and password was assumed by the system to belong to the same user when merging accounts. I don't know if there were any cases of it actually happening, but I remember there being a discussion on it and one of the technical people at WP:Village Pump saying it was "so unlikely to happen it wasn't worth worrying about" or something to that effect, probably trying to hide the flaw per WP:BEANS. But, unlike Wikipedia don't Google and Youtube require email addresses to make an account? Surely two different people wouldn't have the same email address? 82.44.55.25 (talk) 15:29, 2 January 2011 (UTC)
 * You don't have to have the same e-mail address in the above scheme. --Mr.98 (talk) 18:48, 2 January 2011 (UTC)


 * The odds of the same username and the same password seem quite astounding, even given the fact that a lot of people use bad passwords (e.g. 1234, password, qwerty, asdf, etc.). Even if we assume both jsmiths use bad passwords, they both have to use exactly the same bad password. There are a lot of bad passwords out there — not enough to avoid a dictionary attack, but enough to probably avoid accidents. If this is something that doesn't happen 99.99999% of the time, is it worth worrying about the few dozen people it affects? --Mr.98 (talk) 18:48, 2 January 2011 (UTC)
 * Ultimately, this "flaw" is tantamount to a brute force password-guessing attack. The probability of having  identical username and password are the same probability as guessing the user-name and password: and this is an even lower probability than guessing the password alone (i.e., a conventional "brute force" attack).  If the users have weak passwords, then that is the vulnerability; and that account would have been even more susceptible to a random brute-force password-guessing attack than a simultaneous (account-name + password) guess.  So far, there are no effective strategies for precluding brute-force attacks; see our section on countering them.  Furthermore, the flaw in relying on Google for securing documents is really one of social-engineering: Google may be an untrusted party; at best, they provide technical measures that secure the data you communicate to them during the transaction (to prevent a man-in-the-middle attack).  But unless you actually trust Google, relying on security-technology is moot; the data is under the control of a third-party organization (Google); they may share it with anyone they choose (including the random user who happened to have the same account-login-name as you).  Nimur (talk) 20:03, 2 January 2011 (UTC)
 * Hmm, I've been thinking about this and I think there may be the chance of a few clashes like that. I saw some statistics about weak passwords somewhere so I'll need to look at that first to really confirm. Dmcq (talk) 22:09, 2 January 2011 (UTC)
 * Yep it is practically bound to happen. If 'password1' is used by 0.22% of users then about 1 in 500 will choose that. That means it only requires about 250,000 user accounts with similar names joined up and you've a good chance of a clash using that password. Never mind all the people who use their first name as a password for instance. Dmcq (talk)
 * That's 250,000 users with identical usernames. It seems a stretch to me.  But it's the birthday paradox so the number is way less than 250000.  Taemyr (talk) 14:28, 4 January 2011 (UTC)
 * If I am reading this correctly, yes, that would need 250,000 users with identical usernames, which would not be possible, as pre-merger YouTube would only allow one user account with that name, and Google (pre- and post-merger) also only allows one user account with that name (and treats all user accounts with dots in them as the same account). The probability of this happening by chance is infinitessimally small. -- KägeTorä - (影虎) (TALK) 14:54, 4 January 2011 (UTC)
 * Actually it means 250000 pairs of users that has the same name on google and youtube. So no need for all to have the same name.  OTOH, 500000 total users involved.  Taemyr (talk) 09:44, 5 January 2011 (UTC)

Computer user interface and grammar of commands
In English, French, Spanish and presumably most other languages, you see commands worded as instructions (but sometimes the words a/the are inferred). Examples: "Change font size", "Create new document", "Cancel", "Accept" "Execute this, Execute that..."

Why do Arabic language systems display commands in the awkward grammar tense such as: "Changing of the font size", "Creation of a new document", "Cancelation", "Acceptance", "Execution of this and that..."

If you don't speak Arabic, I'll tell you this much, that is not the more intuitive or usual tense used to give instructions. They should be the same tense as used in those other languages. So why are user interfaces for Arabic software written like that? I have seen this also in mobile phones, cameras, and other computer-like electronics. --Roberto75780 (talk) 14:42, 2 January 2011 (UTC)


 * Probably because they were "translated" with machine translation software and not professionally. Either that or you don't know Arabic as well as you think (no offense, that's just the greatest alternative option I can think of). ¦ Reisio (talk) 16:23, 2 January 2011 (UTC)
 * None taken but its my first language and the language I was originally educated in (don't let the name throw you off). --Roberto75780 (talk) 21:16, 2 January 2011 (UTC)


 * This might sound like its a million miles away from your question, but please bear with it for a few moments. I'm Irish (ROI) - Ireland was once a British colony (Not getting into the rights/wrongs on the computing helpdesk). Under a colonial power, there was a massive effort to Anglicise placenames etc. (I come from Wexford - I believe that term is derivative of a norse term for "something<?>_fiord". The Gaelic name for the same place is Loch Garman (Which I understand to mean Garman's Lake) - Point being the Anglicised term was very far from the term that had previously been accepted.


 * How does this relate to your query? - I believe (although I could be very wrong) that the interfaces you are talking about were designed with an Anglophile/western usage in mind, and it is probably a case of poor translation. By this I mean, poor translation of the interface terms rather than proper localisation of the interface. Does that sound like a reasonable possibility? This article (lots of spammy adverts on it) points to the difficulty of arabic localisation. Don't know if its accurate, but might be worth a read.


 * <<>> Darigan (talk) 16:33, 2 January 2011 (UTC)


 * You might like to read Internationalization and localization to learn how commercial software is typically "ported" to another language and locale. Usually, the process is something like this: first, a programmer writes software to perform some function, like "Save File". A user interface programmer then creates a dialog-box, and creates a button in it that says "Save File." Next, an internationalization and localization engineer (who is sort of like a programmer) goes through the user-interface code (note that somebody else designed it, and somebody else wrote it). The internationalization engineer will see the text message "Save File" and replace it with a tag, like SAVE_FILE_BUTTON_TEXT; then they will create an indexed text-file with a series of tags and their text-string as it will appear in the interface. So you'll have a text-file with a list of software-feature tags, and a list of "human-readable strings" that represent them: take a look at this example from IBM. Now, this lookup-table file gets sent to linguists in a bunch of different locales; their job is to guess what the user should be reading when the programmer presents them with a particular software feature. In some cases, this can be very abstract - especially if the software feature is a bit complex. (Try to imagine explaining the purpose of an otherwise opaque bit of user-interface software code without using the String that actually represents it)! (I'm scanning through the menus in Firefox - think about translating "find" in a web-browser, without knowing the context. That could mean "search the internet", "scan the bookmarks", "locate this text snippet within this page", and so forth. The point is, a translator (who often is not a programmer) must scan through thousands of these tags; for each one, they have to use a tiny bit of English text fragment to guess what the software is doing, and then create a localized language representation of the feature. The process is really difficult. (Try reading the official instructions for translating and localizing Firefox!) And take a look at here's a bit of Firefox's Browser Preferences translation table into Norwegian. Pretty tough to guess what the msgstr should be, even if you are a fluent speaker of both languages! (Compare an English version)...  Finally, as a closing note, consider a proprietary, non-free software project, where the translator does not have permission to ask the software developers about the meaning of particular software features (for proprietary intellectual property protection)!  Nimur (talk) 20:30, 2 January 2011 (UTC)


 * I think they probably are just trying to go with a common thing of not giving direct commands as that can be demeaning or rude. So they say what you can do by pressing the button rather than putting you in an uncomfortable position. Dmcq (talk) 21:23, 2 January 2011 (UTC)

urls
I'm trying to use the windows version of grep to extract urls from several thousand files. I don't understand "regular expression" but basically I want it to list all urls that are "http://example.com/123/*" with * being wildcard. How do I do this? 82.44.55.25 (talk) 20:19, 2 January 2011 (UTC)


 * Regular expressions are much more complex than simple wildcards. They have a special syntax; and it so happens that in your example (because the wildcard is at the end of the word), the correct way to search would be to grep on http://example.com/123/ ... without an asterisk.  If that's all you want to do, you don't need to learn more sophisticated regexp syntax; but you should know that (1) reg-exp syntax is very powerful; and (2) reg-exp syntax is not usually "easy" or intuitive for novice users (especially if you are used to the DOS/Windows style of wildcard searching with * and ?).  Your command should look like this:

 grep http://example.com/123/ [FILE(S)...] 
 * Here is the official GNU grep manual, if you want to learn how to search for more complex patterns. Regular expressions are very powerful pattern-matching tools, and can be extremely configurable for complex text patterns.  Nimur (talk) 20:47, 2 January 2011 (UTC)


 * Thanks! 82.44.55.25 (talk) 21:10, 2 January 2011 (UTC)

hjsplit
i download hj-split but i cant figure out how to use it. can anyone help? — Preceding unsigned comment added by Tommy35750 (talk • contribs) 20:49, 2 January 2011 (UTC)


 * Are you using the old school command-line version or the newer GUI version. Here's a link to the manual for the GUI version:.
 * (By the way, I took the liberty of changing the title of your question to "hjsplit".) 93.95.251.162 (talk) 14:27, 4 January 2011 (UTC) Martin.