Wikipedia:Reference desk/Archives/Computing/2010 August 14

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Computing desk
< August 13 << Jul | August | Sep >> August 15 >
Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


August 14[edit]

RSS[edit]

Do online rss readers like Google Reader update feeds even when you're not logged in? So for example, if I added a feed, logged out and logged back in a week later, would google have saved all the news from that week? 82.44.54.4 (talk) 00:00, 14 August 2010 (UTC)

In the specific case of Google Reader, it would appear so. I subscribe to BoingBoing there, but I never actually bother looking at Google Reader. When I opened it just now, it shows >700 entries for that feed. But if I look at the actual RSS than BoingBoing syndicates, they're only publishing 30. So Google Reader must have been periodically visiting the feed and keeping its own database updated. That's a sensible thing for it to do, but it's a decision that the Google Reader people have chosen to make. They could quite reasonably have stopped visiting after a while, or kept only so many entries (maybe they have, it's hard to know). Other online readers may work entirely differently. -- Finlay McWalterTalk 00:31, 14 August 2010 (UTC)
Having gone through a longish (several month) period without checking Google Reader, I can hopefully add some perspective. *Most* of the feeds were maintained in my absence - that is, as far as I can tell, when I checked back in all the updates that were added when I was gone were there. However, some of the "personal" feeds (specifically "keep me updated on these search term"-type feeds where I was the only subscriber) *weren't* updated, and it looked like I was missing portions. Here's how I think Google Reader works: Google only keeps a single list of items for each feed address, and that's shared by everyone who subscribes. That's why you can get several years worth of feed history when you sign up to a feed - Google has the information stored from earlier subscribers. However, Google only updates the feed if *someone* is actively looking at it. It doesn't have to be you, it could be someone else who has subscribed. If it's a popular feed (say the BBC News feed), someone else will probably cause it to be updated while you're gone. However, if you are the only one watching it, Google won't bother to check for updates unless you are logging in, so you may lose some if you don't log in regularly. That's Google Reader - I don't know about other online feed readers. -- As a final note, RSS feeds usually contain more than just the last entry/last day's entries (some rarely updated ones can even contain the complete history). So even if Google doesn't check the feed for several days/weeks, there is a chance that older posts will still be in the feed itself. -- 174.21.233.249 (talk) 20:27, 15 August 2010 (UTC)

Identical[edit]

This is probably a silly question, but I just want to clarify anyway; is anything lost or altered when files are copied? For example, say I have a file on my computer, I put it onto a usb drive and gave it to a friend, he/she then uploaded it to rapidshare where someone in Australia downloads it, burned it to a disk, and sent in back to me in the post. Would the file on the computer and the file on the disk be identical? 82.44.54.4 (talk) 00:05, 14 August 2010 (UTC)

Yes, they should be identical - that is, if the file content was different, that would be considered a defect in the transmission process. There are some (mostly nitpicky, won't matter to you in reality) issues. Firstly, the meta information (the file's name, modification-date (and other date info that some computers store about files) and owner info might be changed or lost) - mostly you don't care about that, but sometimes if you're transmitting files where that does matter, artefacts of the transmission process can break things you didn't expect. Secondly some file systems store additional data with files (NTFS stores "alternate data streams", MacOS stores "resource forks", and some fancier filesystems store complicated things like revision info); depending on the tools you use to transmit and remotely-store such files, these data might also be lost - this too is a rare and esoteric thing, so you'll very very rarely care about this. And lastly there are weird doings with text files (truly .txt ascii files) where transmitting from a DOS/Windows system to Unix (or to Mac) might cause the line termination characters to be translated to suit (see Newline for a painful explanation) - sometimes you'll want this to happen, sometimes you definitely won't, but these days most programs handle everything as binary and you won't see any difference. So, for a simple binary store-and-retrieve thing like Rapidshare, the files should be 100% identical - you and your Australian counterpart can both generate md5 checksums of the file and you should both generate the same code. -- Finlay McWalterTalk 00:17, 14 August 2010 (UTC)
Depending on your level of interest, you might find the article on Parity bit interesting. Vespine (talk) 00:57, 14 August 2010 (UTC)
Not sure if you're asking out of pure curiosity or you're planning on copying a lot of files. If it's the latter, I've found data transfer to CDs and Flash media is less than 100% reliable, particularly at fast write speeds for CDs and old or cheap USB flash drives. Generating checksums as Finlay McWalter mentions is a great idea to be sure the transfer went okay, especially with photos and other irreplaceable files. To be clear, digital files don't "degrade" with each transfer like you would expect from a cassette tape, but every once in a while some bytes can be copied wrong and corrupt or distort a file. Checksums alert you when this happens so you can redo the transfer.--el Aprel (facta-facienda) 04:10, 14 August 2010 (UTC)
While I usually write checksums to optical media myself, note that programs like Nero and Imgburn can verify the written image or files after burning Nil Einne (talk) 07:29, 14 August 2010 (UTC)

Activating OpenType Features[edit]

How do I activate the various "features" of an OpenType font in a program such as Word 03? (And if I can't, what program can I do it in?) For instance, this page (http://robert-pfeffer.spacequadrat.de/schriftarten/englisch/nachgeladener_rahmen.html?pfeffer_mediaeval.html) mentions things like "activating the 'hist' feature for historical text layout" and "activating stylistic set 'ss01'", but I have no idea how to do that. The "OpenType User Guide" isn't very helpful either. Help would be appreciated. 64.179.155.63 (talk) 06:13, 14 August 2010 (UTC)

The program in question (e.g. Word) needs to support the features. You can't make it support them if it doesn't already do so. I don't know if Word '03 does. This table (at the bottom) shows how some of these features are supported in various programs — Word 2010 does OK, for a word processor. At the bottom of the page it shows what it looks like in a few programs that support these kinds of features. --Mr.98 (talk) 19:31, 14 August 2010 (UTC)
But how do I activate those features? Is it usually in the Format->Font window or.. what? No website I've found says anything about how to do such cryptic, specific actions... only about what programs can do them. 64.179.155.63 (talk) 00:35, 15 August 2010 (UTC)

Defaulting Australian English in Word 2007?[edit]

Hi everyone,

does anyone know how to make Australian English the default language in Word 2007, and actually remove the US, so that any US spelling will be highlighted and not slip through with the Australian?

Thanks in advance Adambrowne666 (talk) 10:03, 14 August 2010 (UTC)

As far as I'm aware Word 2007, as with all versions of Word should highlight spelling considered incorrect in Australian English (even if okay in US English) by default when the document is set to Australian English. I've tried it myself on Word 2007 and it seems to work. Is your version not doing this? Bear in mind there are some words where it's not clear if an alternative spelling is considered incorrect in Australian English (as with other variants of English) so the Word 2007 may have these in their Australian English dictionary and they aren't going to be considered incorrect. For example, recogise/recognize appears to be one of these (as is mum/mom but not sceptic/skeptic). Colour/color is a good test if you're looking for one.
You may not agree with all of Microsoft's views on what's correct/incorrect in Australian English (and some of them may be more 'bugs' then intentional choices anyway) but there isn't AFAIK any way to be 'stricter' or something so your best bet is to manually remove words you consider incorrect from the Australian English dictionary (well there's probably some way, I don't know how) and complain to MS. You can also add autocorrections, Word already does this for things like kilometer (but not meter for obvious reasons).
In terms of how to change the default, click on the language shown at the bottom of the screen and you should get a list, select English (Australian) and then click on default and ok and it should change the default. Note that this does indeed only change the default. If you have any existing documents where the language is set and saved (it usually is with Word documents) it will not affect them. You will have to manually change the language in existing documents or use a macro or something to modify all existing ones. You could similarly probaby make a macro which will automatically change documents you open to be Australian English I'm not aware Word has such a feature itself.
BTW, AFAIK, Word follows your OS language when it's installed, so you may want to change your OS language if it's something else to be Australian English for this and other programs.
Nil Einne (talk) 10:46, 14 August 2010 (UTC)
Thanks for the thorough answer, Nil Einne - I was also sent this by a friend step by step Adambrowne666 (talk) 03:38, 16 August 2010 (UTC)

McGill Email[edit]

Hello. Is it possible to check my McGill email through Windows Live Mail? (Outlook Web Access would be the third party if I checked my email through my web browser.) The second setup window asks for my incoming and outgoing server information. Thanks in advance. --Mayfare (talk) 12:05, 14 August 2010 (UTC)

A quick search for 'mcgill email' leads me to [1]. Clicking on "Email and Calendaring" then "Email for Students" leads me to [2]. Under the "Accessing McGill email" section there is a "Configure an email client (application) to access McGill Exchange" link which leads to [3]. There are a bunch of stuff there telling you how to configure it for various clients including Windows Live Mail. If this is the wrong McGill, perhaps it would be wise to specify which McGill you are referring to in the future. However generally speaking the people who are able to best tell you if you are able to access your e-mail via an e-mail client would be the people who provide you the e-mail. They also tend to provide you the server settings and in fact very commonly as in this case with specific instructions for popular clients. POP and IMAP are the most commonly used protocols for accessing email by standalone clients so a search for your email providers POP or IMAP servers is often a useful test. Nil Einne (talk) 14:10, 14 August 2010 (UTC)

Changing default printer does not stick[edit]

I'm running XP Professional - Service Pack 3 with automatic updates fully up to date. When I change the default printer - from a colour inkjet (HP Deskjet 5652) to a laser (HP Laserjet 1020) - the laser only stays the default until I reboot, then it reverts to the inkjet by itself. Both are connected by USB. How can I make sure the change becomes permanent? Roger (talk) 12:56, 14 August 2010 (UTC)

Random idea - swap the two connectors over in the USB slots? Maybe your PC assigns defaults based on the order they are connected in. Exxolon (talk) 16:46, 14 August 2010 (UTC)
You could also try deleting the Deskjet, rebooting, and re-adding it. From a quick google search, it doesn't look like other people with the problem have found simple solutions. Hmm. Indeterminate (talk) 20:47, 14 August 2010 (UTC)
I once used a library computer which was set up so that it created a new user from default settings every time someone logged in and then deleted that user when they logged out. It was good for the library so that any one user didn't mess up the machine's configuration, but if you had that kind of thing going on at home or at work it would be damn irritating if you wanted to change the default printer because that setting wouldn't stick. Astronaut (talk) 07:47, 15 August 2010 (UTC)
Thanks for all the advice. Here is how I fixed it: 1. Unplugged both printers. 2. Removed the Deskjet from "Printers and Faxes" 3. Set the Laserjet to default. 4. Rebooted. 5. Reconnected both printers. This caused the Deskjet to be re-installed by the "Found new hardware" function. 6. Checked that the Laserjet was still default. 7. Rebooted again. Problem solved! Roger (talk) 15:52, 16 August 2010 (UTC)

UDP[edit]

I read the UDP article but I don't get it. Can UDP be used instead of TCP? What is UDP for? What programs use it? 82.44.54.4 (talk) 14:06, 14 August 2010 (UTC)

Simply put, TCP and UDP are two different transmission protocols for moving packets across an IP (packet switched) network. One distinction between TCP and UDP would be their connection orientation -- whereas TCP is a connection oriented protocol, meaning that an session is opened between client and server (and this session is opened, closed, acknowledged, etc) -- UDP is a connectionless protocol. Another distinction between TCP and UDP is that TCP is a "reliable" protocol and UDP is not. This means that when a client sends a packet using TCP to the server, it can know for sure whether or not that packet arrived there. In UDP, there is no acknowledgment of packet receipt. Here is a good breakdown —Preceding unsigned comment added by Rocketrye12 (talkcontribs) 15:00, 14 August 2010 (UTC)
(ec)They do rather different things, so mostly you wouldn't use them for the same job. The advantage of TCP is that it's "reliable", which really means that the packets are delivered to you in the order they were sent. So if a packet is lost or damaged, TCP will resend it, and won't deliver to the receiver any subsequent packets until it's got a good copy of the damaged one. So if reliability is what you care about, then you use TCP. Web pages and images are sent (over HTTP) over TCP, because a web page with the middle missing, or an image with the top half missing, is useless. But that reliability has a cost - if a packet has to be retransmitted, all the subsequent packets pile up behind it. While you're waiting for the retransmission, the delivery of packets appears to stall. For stuff where you care about the timely delivery (often more than reliability) using TCP would cause jumps. So things like video-chat use UDP instead - that way a lost packet doesn't hold up all the subsequent ones. Such a protocol thus has to be tolerant of such a fault - so for video it's okay if the picture skips, but it mustn't stall or disintegrate when a packet is lost. So mostly TCP and UDP do different things, and you'd use the one appropriate to your application. A few applications use UDP for reliable transfer (I think edonkey does), but that means they have to build their own retransmission stuff into their application protocol. -- Finlay McWalterTalk 15:04, 14 August 2010 (UTC)
Strictly, I've conflated "reliable" (you know the packet gets through) with "in order" (the packets arrive in the order they were sent) above; most applications either need both or neither. UDP is neither, TCP is both. P2P protocols like edonkey are unusual in that they have a meaningful use for "reliable, not in order", something that neither protocol can give them (so they essentially implement their own). P2P systems are very fault tolerant and deal with counterparts that only have part of the data (you can download a file from a bunch of sources, where each individual doesn't have a whole copy, so long as each byte of the image is held somewhere by someone in the swarm). So P2P clients are willing to build up a piecewise mosaic of a file, with the pieces arriving in crazy orders from a bunch of places. -- Finlay McWalterTalk 15:10, 14 August 2010 (UTC)
Agree with all that has been said above. Just to address the OP's original question (in case he/she has been lost in technical jargon): "Can UDP be used instead of TCP?" Well, if you are the user of a program, no; there are very few programs that let you "choose" between the two options. It is very rare for any program to offer a "mode" or user-configurable option to switch between the two. This is because (as has been explained above) the program is designed to use the protocol that best suits its needs; TCP and UDP function differently, and provide a different "contract" regarding data delivery. Just "switching" transmission protocols would be trivial, but the program would probably break in weird ways if it's using a scheme other than the one it was designed for. So it would be useless to offer this as a "menu-option" to the user. If you are the designer of a program (e.g. a programmer), you can easily switch between UDP and TCP, as long as you program the system to properly handle the resulting data-transmission behavior. Typically, a programmer does this by selecting either a TCP or UDP socket API in the programming-language of their choice; in the special case of C or assembly-programming, you can actually roll your own protocol-level data management and "re-implement" either protocol at the device-level. Nimur (talk) 05:59, 15 August 2010 (UTC)

Windows Movie Maker[edit]

Whenever I try to use this, it tells me I have missing "codecs". What codecs should I download please? And also, what do I do with the codecs once I have downloaded them? I have WinXP Sp3. Thanks 92.28.247.227 (talk) 20:57, 14 August 2010 (UTC)

What this will mean is that you are trying to open a movie file in a format that Windows media player doesn't support out-of-the-box. Codecs are library files that allow applications to work with more file formats. In this case I would have a look at free-codecs.com for a codec matching the file format you want to work with. Rjwilmsi 21:12, 14 August 2010 (UTC)

Thanks, but I'd like some more information if I may. Will all and any of the codecs work with Windows Media Player? Are some better than others? Why are there so many different ones? Should I have more than one codec pack installed? Are they all compatible or incompatible? How can I tell what I already have installed already? Is there any kind software program that can tell me what codec packs I should install? What do I do with the codecs once I have downloaded them? Thanks 92.28.248.196 (talk) 20:16, 15 August 2010 (UTC)

GSpot is a free program that will analyze a given video file and determine whether you have the proper video and audio codecs installed. It is small and quick and helpful. There are countless codecs you could install, but in practice, what you need most is DivX (also free, though be careful what options you choose during installation). Matt Deres (talk) 15:40, 17 August 2010 (UTC)