Wikipedia:Reference desk/Archives/Computing/2007 February 22

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.

February 22

WiFi for Linux

I hope I placed this question right, feel free to move it around.
I just started using Ubuntu linux, and have come across a bit of a problem: WiFi. I tried using LinuxAnt DriverLoader, it didnt work, I tried NDIS Wrapper, it didnt work. I tried just using what linux had to configure, it didnt work either. I am either doing something wrong, or need to find another way to add this. How can I get WiFi to work Linux? I have the latest kernal, and I have a USB WiFi Adapter. Zylstra555 05:03, 22 February 2007 (UTC) --user:Zylstra555 9:02PM Feb 2007

The best place to ask your question for this would be at http://www.ubuntuforums.com/ or alteratively #ubuntu on irc.freenode.net. 61.25.248.86 00:26, 23 February 2007 (UTC)

Stability of Firefox + Add-ons

If all the add-ons for Firefox are coded by third-party amatures, aren't they likely to conflict with each other and other components of any given computer system and likely cause more problems compared to Opera where the entire thing is coded by a single development team and subject to more rigerous testing? --Seans Potato Business 00:37, 22 February 2007 (UTC)

What type of add-ons are you talking about? Firefox without extension is a fully functional web browser. Few extensions have major issues, and the popular ones are generally rock solid. I typically use Unplug, Adblock Plus, and Greasemonkey, and I have had no stability issues. I'm not sure that you can say that Firefox extensions receive less testing either, or that the creators are necessarily amateurs. -- Consumed Crustacean (talk) 00:46, 22 February 2007 (UTC)

Add-ons generally do not conflict with each other - although that is a possibility. Keep to the addons.mozilla.org extensions which are checked by volunteers before being made available to the public. I've looked at the code for some of the popular ones and they've all looked well-coded to me. I've only had problems with plug-ins (Java and Flash) which are coded by professionals! --h2g2bob 02:54, 22 February 2007 (UTC)

There are a number of addons that are famous for screwing up the browser. See Mozilla's Problematic Extensions. Oskar 16:30, 22 February 2007 (UTC)

Sharing Single Copy of Program by Two Linux OSs?

In a situtation where Ubuntu and OpenSUSE are installed on the same computer, is it advisable or even necessary that application are installed separately for each or can they be installed once, perhaps on a third partition and used by both operating systems (that way, additions to an OpenOffice dictionary under one OS wont leave the other dictionary without those additons plus loads of other divergent settings that would occur. --Seans Potato Business 02:44, 22 February 2007 (UTC)

SUSE uses rpm, Ubuntu uses deb. They have different package systems and even if you really really really want to do it by sharing some of the key directories between them, what would be the point then for having two systems?--antilived^{T | C | G} 04:20, 22 February 2007 (UTC)

This seems unnecessary (applications aren't that big, OpenOffice is still only about 100 MiBs), but what you could do is a third partition for all of the stuff that does take up room, ie data. You can put your music and movies and what-have-you there. And, as already stated, it would be a bad idea to share software since they use different package managers (I thought openSUSE used YaST, btw?) Oskar 16:26, 22 February 2007 (UTC)

You're looking at a more or less irreconcilable linking/depedency mess. What you could potentially do is share the /home filesystem so that settings are kept, but sharing the actual binaries (especially if they're dynamically linked) isn't really feasible. While it may be possible in certain (mostly statically linked) cases, it will be a big headache and waste of time on the whole. -- mattb @ 2007-02-22T18:53Z

iTunes

Hi. I think I've managed to burn an MP3 Disc in iTunes before but in folders for each individual album. Does anyone know how to do this? I'm sure it's easier with Nero but I can't get it to work.

If you are running Windows XP, you may want to just drag and copy the music onto the burning drive. When asked, select "Make Audio CD" to make a 80 Minute CD that can be played in all Compaq Disk Audio CD Players. If you are not using Windows XP, I would suggest searching through the iTunes help files.

If I'm understanding your question right, the folders you are referring to are playlists. So to make a cd out of them using itunes directly, select your music (all the songs at once) and in the file menu at the top left select to "Create Playlist from Selection" (I think that's the wording). Then the playlist will be created as a folder in that left tab thingy. So you just click on the playlist, insert your blank cd, and click the burn cd icon in the upper right. Hope that answers the question. The above mentioned suggestion should work perfectly fine too, but it requires you to locate the songs on your hard drive. 70.108.199.130 06:27, 22 February 2007 (UTC)

Vista issue

Just got a upgrade version of Vista home basic but because i just got a new hardrive i had to bypass entering the product key because my XP system was not detected. Therefore when i try to activate windows using the upgrade product key it says it can not use the key because of it being an upgrade key. i have basic on my drive right now but i need to figure out if there are any sites where i can purchase a new product key to activate my copy or to find out when new product keys will be avalible. Any help would be good.--Biggie 08:31, 22 February 2007 (UTC)

From what I've read, Vista has tons of 'trip wires' to stop anything they don't like. The upgrade must have a perfectly happy copy of XP existing prior, even though it eventually wipes everything out anyway. There is apparently a way to install Vista twice to fool it, but I don't know the details. --Zeizmic 12:31, 22 February 2007 (UTC)

You can Get a New Product Key by Calling (800) MICROSOFT (642-7676) or you can go out to the store and buy a copy off the shelf but since you already have the disk then you might as well call for a new key. ( to get human Press 0 at each prompt, ignoring messages. ) --Nerdd 15:54, 22 February 2007 (UTC)

Thanks for the advice i'll try the number

Just install Vista again over the already installed copy of Vista as an upgrade, and enter your product key at that time. kmccoy (talk) 20:21, 22 February 2007 (UTC)

IntrAnet naming customs?

Are there any naming conventions for intranets? for example, if it's the intranet for company XYZ, would "Intra XYZ" or "XYZ Intranet" or "Inside XYZ" or "XYZnet" or whatever be appropriate? I realize you can name a site whatever you want, but maybe there is a popular convention that is especially clear or considered good usability.--Sonjaaa 18:43, 22 February 2007 (UTC)

I havent come across any popular conventions. Where I work its the company name and Intranet - I wouldnt have thought it would matter because the only people whoa re going to look at your Intranet is your company staff - no one else would/should have access to it outside the company --PrincessBrat 12:03, 23 February 2007 (UTC)

About 90% of places I have contracted to (~150 clients) use the default windows network name WORKGROUP. The others use something related to the company, like NIKE. It's usually redundant to use "XYZ NET". That's like ATM Machine or, my favorite redundancy, hot water heater. —EncMstr 23:28, 23 February 2007 (UTC)

Dosbox or Wine

Is this hypothetical situation possible? I am browsing files on my Linux computer as usual, and I find a .exe file. When I click it, or type it's name in terminal, either Dosbox or Wine is automatically launched (without me needing to specify which one). Is this a realistic dream? Duomillia 20:13, 22 February 2007 (UTC)

It should be possible. --Carnildo 21:17, 22 February 2007 (UTC)

It's certainly possible when you are browsing around in your favorite file manager. In Nautilus, right click on the .exe file, choose properties, then Open with and you can choose Wine Windows Emulator (which is kinda strange, since Wine Is Not an Emulator). That will open the file in wine any time you double click on it. From the command line it is harder, you can simply just change the permissions of the file and run it, that won't work. There might be a way, I'm not skilled enough in linux to be able to tell you for certain (well, there is certainly some way), but by far the easiest thing will be just to simply type in "wine " ahead of the file. It's five characters, fer chrissakes, deal with the pain! :) Oskar 22:40, 22 February 2007 (UTC)

Email Viruses

How do they work? Do they only work if an attachment is downloaded? If the download is executed? If the email is opened? I don't get how a virus sent by email would work if you didn't download something specific. Thanks, 70.108.199.130 21:34, 22 February 2007 (UTC)

E-mail viruses typically either entice the user to open the payload, or exploit a bug in mail programs like Outlook Splintercellguy 22:03, 22 February 2007 (UTC)

The classic example is an outlook bug which allowed emails to contain unrestricted scripting which activated as soon as you opened the email. That is, when you opened it, a program would run that would infect your computer and send itself copies to every one in your address book. This system was created before people (well, Microsoft) had realized just how important Internet Security really is Oskar 22:57, 22 February 2007 (UTC)

This is a very good question. According to one school of thought, an email virus should be practically impossible. No one, so the theory goes, would ever think of designing an email program that would run arbitrary, unchecked, untrustworthy code received in unsolicited mail from an unknown, untrustworthy recipient. Doing so is the "safe computing" equivalent of rolling around naked having sex with perfect strangers in a garbage- and needle-strewn alley behind a crack house with bad plumbing.

However, the previously-described school of thought had not reckoned on one thing: Microsoft. For the longest time, Microsoft thought that running arbitrary, unchecked, untrustworthy code received from any unknown, untrustworthy recipient was just dandy. After all, if an attachment is a .JPG, you hand it to an image-viewing program to display it, and if an attachment is an .MPG, you hand it to a movie-playing program to play it, so if an attachment is a .EXE, why not just hand it to the CPU to run it? And the computing landscape was changed forever. Not all unchecked, untrustworthy code is malicious; in fact an environment in which unchecked, untrustworthy code is routinely executed makes certain nondestructive actions very convenient. And by now the computing industry (at least, that part of it dominated by Microsoft) has become so dependent on that convenience that the "obvious" solution to the email virus problem, namely utterly disallowing the easy execution of untrustworthy externally-received code, is widely held to be impossible. So instead we all (er, well, most of you all) have to put up with (imperfect) third-party virus protection, and UAC, and stuff.

As Oskar mentioned, the executable-attachment problem isn't even the only problem. At various times, Outlook and certain other email clients have suffered various egregious bugs under which they were willing to run the untrustworthy, unchecked code all by themselves. So while the classic email virus depends on your clicking on or otherwise opening an attachment in order for it to be executed, at times there have been viruses which infected your machine when all you did was view the message, or even when all you did was have the message's header display in your inbox. In particular, Outlook Express was for years a virus-writer's best friend, although I gather it's been mostly fixed by now.

Another sad aspect of the email virus problem is that, too often, we end up blaming the victim. How often have you heard someone say, "The problem wouldn't be so bad if only there weren't so many stupid users clicking on obviously bogus attachments"? But to me, rigging up email clients so they execute an executable attachment with one click, and then attempting to prevent a worldwide computer virus epidemic by asking users to remember not to click on the wrong kind of attachments, is sort of like equipping all cars with ejection seats, with the "eject" button right in the middle of the steering wheel next to the horn, and trying to solve the resulting accidental-ejection epidemic by putting up conspicuous billboards along busy highways reminding people to be careful in finding the horn button if they need to honk. But what do I know. —Steve Summit (talk) 01:47, 23 February 2007 (UTC)

Your argument doesn't take the user into account. The scenario is more accurately described as:

Microsoft: Here's an email program. You can only view text. If you want to see the attachments, you have to save them to your disk and open them with another program.

User: I'm too stupid to save a file to a disk. Let me just open it with a program.

Microsoft: OK. You don't have to save it to a disk now. You can open it directly from the email.

User: I'm too stupid to select a program. Just open it for me.

Microsoft; OK. You don't have to select a program now. Just say you want to open it and it will open with the most common program for the file type.

User: I'm too stupid to ask it to open a file. Open it for me.

Microsoft: OK. It opens all the attachments without asking you now.

User: I got a virus! Why did Microsoft do this to me!?!?

So, now Microsoft is the bad guy because users have spend the last 10 years complaining that they simply too stupid to use computers. In my opinion, they should have kept it the way it started and told the users to use their brain just a little. It wouldn't kill them. --Kainaw ^(talk) 03:27, 23 February 2007 (UTC)

But there's a middle ground, which for some reason is never pursued. I'm with you -- really with you -- on those attachments and the lazy users, and on the appropriateness of near-automatically opening the data attachments in the associated viewing apps. But where is it written that the thing to do with an executable attachment is to give it raw, unfettered access to the entire CPU, i.e. to just run it, i.e. to treat it not as data to be handed to a program, but as a program itself? That's the wickedly unsafe thing, and I don't think those lazy/naive/stupid users you're speaking of would have clamored for this easy executability of code. —Steve Summit (talk) 03:39, 23 February 2007 (UTC)

Change Date/Time Field in GMail

Is there a way to change the date/time field in emails sent from GMail, so that email appears to have been sent from whatever date you specify? If not, is this possible in any email client? Thanks. Daniel (‽) 22:19, 22 February 2007 (UTC)

This is currently not possible in GMail, nor any client I know of. However, if you were to custom-code an email script in, say, PHP, this *may* be possible by changing the headers. This is complicated stuff and unless you're really determined you won't get very far! JoshHolloway 22:31, 22 February 2007 (UTC)

The orig-date is a required header (or so RFC822 tells me), so in theory you could absolutely change it. If it will have any effect, I don't know, the date displayed might very well be the date that your email-server received it or when your client downloads it (although that would not be very smart). You'd have to run some experiments with various email-servers and clients to find out if you really could send your friends emails from the future ("The future, Conan?", cue "In the year 2000" music). Oskar 22:50, 22 February 2007 (UTC)

[edit conflict]

In my experience, custom-coding email is neither complicated nor particularly difficult, though I suppose it's not for the faint of heart. On most Unix and Linux systems, at least, you can construct a raw email message (typically with at least To: and Subject: headers, a blank line, and the body of the message), and pipe it into /usr/lib/sendmail -t, and off it goes. Now, the man page for sendmail says that it "is not intended as a user interface routine; other programs provide user-friendly front ends; sendmail is used only to deliver pre-formatted messages". Maybe I'm crazy, but I find manually preformatting messages so easy that I do it that way all the time. (Telnetting to port 25 and doing SMTP by hand, as TotoBaggins demonstrates below, that's a little trickier! :-) )

Besides the To:, Subject:, and Cc: headers you supply, certain other headers, such as From, Date:, and Message-ID, will be automatically added, although if you supply your own, they're accepted. So as long as you get your own Date: line in one of the formats stipulated by RFC822, you can set any date you want, and in my experience the date you so set is the date that's presented to the recipient. (If anybody wants me to, send me your email address and I'll be happy to demonstrate.)

Beware, though, that a Date: line more than a few hours in the past or future is for some reason a hallmark of spam, and many spam filters will give such messages a very high spam score, so they're pretty unlikely to get through to an arbitrary recipient after all.

Also, before someone scolds me, I suppose I should say that just because you technically can forge just about any part of an email header you want to, doesn't mean that you should. Forgery with intent to deceive is Wrong. —Steve Summit (talk) 00:57, 23 February 2007 (UTC)

You can easily tell the mail server the wrong date, and most receiving clients will happily display it, probably on the grounds that the sending mail client knows better what time the message was sent than any MTA along the way. The following SMTP session produced a mail that did indeed display as coming from the future. --TotoBaggins 00:38, 23 February 2007 (UTC)

$ telnet bag-end 25
Trying 127.0.0.1...
Connected to bag-end.
Escape character is '^]'.
220 bag-end ESMTP Postfix (Ubuntu)
HELO sean-desktop
250 sean-desktop
MAIL FROM:<gollum@oz>
250 2.1.0 Ok
RCPT TO:<toto@bag-end>
250 2.1.5 Ok
DATA
354 End data with <CR><LF>.<CR><LF>
Subject: Bad Date Test
Date: Thu Feb 22 19:22:22 EST 2017
To: toto@bag-end

Hello from 2017!
.
250 2.0.0 Ok: queued as A760B1C564
QUIT

Thanks all. To summarise you learned opinions, it can be done, but not without more knowledge of computing than I have(. Never mind. Daniel (‽) 21:41, 23 February 2007 (UTC)

Turning off Norton Pop up Blocker

I have a new computer and when I open up a certain window in this site for my work it will only allow me to do so the first time. After I save and exit I try to go to the next link it won't allow me to open any links again instead it gives me a sound and says "a pop-up window was blocked." Sometimes if I try long enough and I exit that site and re-enter it will let me or open a pop up window but then it won't let me again. I also tried going through the other way turning down the security level and clicking off the check mark it still won't allow me to open up pop-up windows from that site. Also a side note is that my computer allows pop up windows on other sites. I believe it is Norton Virus program, so how can I turn it off temporarily, or at least the pop up blocker for that program?

Some googling turns up this site (clearly in danger of exceeding URL length specifications) which gives you a short little instruction on how to do it. Does that work? Oskar 22:54, 22 February 2007 (UTC)