Wikipedia:Reference desk/Archives/Computing/2008 October 14

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.

October 14

Making a different email client the default in IE

Hello all. I want to have my computer's default email be myname@verizon.net. When I click on "contat us" on amny sites that attempts to openb up an email screen but it openbs up to a default ermail client. I know where to access this to change it: I go to properties in Internet Explorer, and in there, it has a place to change the default. But, unfortunately, in the drop down menu, it gives a host of different services , such as AOl and hotmail. I do not see any way to fill in a new default (like I said, I want it to be my verizon email). Can you please advise how to manually change the field to fill in and remedy this problem. Thank you very much in advance.--68.237.217.229 (talk) 04:28, 14 October 2008 (UTC)

You need to change how Internet Explorer handles mailto: links. I do not have Internet Explorer handy for me. Under Mozilla Firefox, all you need to do is Options > Applications >> mailto and change the setting to desired one. Is there any way for me to convince you to test drive Mozilla Firefox as a replacement for Internet Explorer? Thanks. Kushal (talk) 08:44, 14 October 2008 (UTC)

I use firefox pretty much exclusively, and when I'm not, I'm on Safari. This is for a person who is computer illiterate who I would have to train in all new ways if I changed the browser she uses. This "problem" is only a problem if you don't know, for example that you can simply highlight the address copy and go to your own email client. I would be much easier for me to change the default, trust me on this. Can someone help me actual change the email default?--71.249.111.130 (talk) 21:12, 14 October 2008 (UTC)

Windows XP SP2 should allow you to set program access and defaults. 69.150.163.1 (talk) 02:19, 15 October 2008 (UTC)

The original poster is confused why AOL and Hotmail appear in the list of e-mail clients, but the verizon.net e-mail provider can't be easily added. You can't just add arbitrary websites or e-mail providers to the list. A program must be installed for it to appear in this list. For example, to add Yahoo Mail to that list, you have to install the Yahoo Toolbar for Internet Explorer. To add Gmail to that list, you have to install the Gmail Notifier.

For a verizon.net address, I couldn't find an answer for certain, but here are some possibilities to try:

• You might try installing the Verizon Broadband Toolbar, but I'm not sure if it adds verizon.net mail to the list of default e-mail programs.
• You can go to activate.verizon.net, select "Configure additional PC or reinstall software", and select to customize the install. There might be a mailto plugin available as one of the things you can install.
• Another alternative is to set up Outlook Express to use the verizon.net address, then select Outlook Express in the list as the default e-mail program. This will allow you to use the verizon.net e-mail address, but will look and feel different from using the verizon.net website.

--Bavi H (talk) 05:26, 15 October 2008 (UTC)

Excel VBA query

value in cell "B15": "D" & rwcnt & ":" & "J" & rwcnt

when we read the text in cell 'B15' through VBA code, the code should consider rwcnt as a variable and substitute its value so that the whole text in cell 'B15' becomes a range, like: D6:J6 (if rwcnt=6)

How to do this? —Preceding unsigned comment added by 157.191.2.16 (talk) 07:26, 14 October 2008 (UTC)

You want cell B15 to contain the text "B6:J6"? If so,

Range("B15").Value = "D" & rwcnt & ":J" & rwcnt

works for me. I declared rwcnt as an integer. Zain Ebrahim (talk) 08:00, 14 October 2008 (UTC)

rwcnt is not constant and keeps on changing its value, so we put the text "D" & rwcnt & ":J" & rwcnt

and as the value of rwcnt changes, on accessing the value of cell B15 ie. "D" & rwcnt & ":J" & rwcnt, we can refer to the range contained in B15

eg:

rwcnt=10, then set rng=range("B15").value so this must set rng to (D10:J10) —Preceding unsigned comment added by 157.191.2.16 (talk) 08:21, 14 October 2008 (UTC)

I'm not entirely sure what you're doing but storing a range as text in a cell seems inefficient. You should just reference the range directly for whatever you need it for. Maybe if you told us what you're trying to achieve we might be able to help. Zain Ebrahim (talk) 09:47, 14 October 2008 (UTC)

Secure Password

If I choose a sequence of letters like "asdfghj" -following a path on the keyboard - for my password, is that secure? What about choosing the first letter of every word in a sentence like 'wtfe' for "Wikipedia, the free encyclopedia". Besides being too short, is there any problem in the second approach? Mr.K. (talk) 11:01, 14 October 2008 (UTC)

Short passwords are vulnerable to dictionary and brute force attacks. They should be avoided at all costs. Something as short as four letters is also vulnerable to rainbow table hash cracking in a crazy way. Avoid it! "Besides being too short"—that's the be-all and end-all of this particular limitation.

Long passwords are as secure as any other password. They aren't vulnerable to brute force attacks but there are a million other ways they can be compromised with poor user habits or poor technical setups. --98.217.8.46 (talk) 11:27, 14 October 2008 (UTC)

The best passwords involve alphanumeric characters and numbers along with capital and lower-case letters. Here are some randomly generated suggestions from my school's "Change your password" webpage: =HogTwIT2, 4LaB*smuG, %Dug2sPEc, 8*JErkteA, CATbanG5#, jumP}9Yea, moAN&6LiP.--droptone (talk) 11:41, 14 October 2008 (UTC)

• I'd wary of the first approach, as most "dictionary" techniques should include obvious combinations. The second is certainly better, as it is effectively using the sentence/phrase as a mnemonic, and there's nothing particularly wrong with that, so long as the combination itself can't be easily guessed. However, for better strength (as Droptone mentioned) you should probably be combining case and numerals as well, with other characters being cool, too - thus the sentence approach may be limiting in this regard. Perhaps "Wikipedia is the 01 Encyclopedia" giving Wit01E, or something similar, will be more effective. - Bilby (talk) 11:46, 14 October 2008 (UTC)

@98.217: I was not suggesting my password could be so short. I was only exemplifying a concept.

@droptone: "alphanumeric characters and numbers"? Are you sure that you didn't mean "alphanumeric characters and letters and numbers"? I see that there are real words in your suggested passwords: lab smug, dug, jerk tea, cat ban, jump. Would it compromise the security of an encrypted file?

@bilby: couldn't a robot scan for sentences, pick the first letter and build a dictionary of passwords?

Perhaps the best approach is to open 20 pages, takes a random letter and mix it up with random special characters...Not easy to remember... :(Mr.K. (talk) 12:04, 14 October 2008 (UTC)]

Yes, except that the range of possibilities for sentences is much greater than the range of possibilities for words. A seemingly random string of letters that happens to be part of a mnemonic is going to be just as good (to all practical intents) as a random string of letters that isn't part of a mnemonic.

It is part of why there is a push towards passphrases for security, instead of passwords - it makes dictionary attacks much harder, while still making it easier to remember - although the mechanism for brute force attacks changes, rather than is broken. At the same time, obvious pass phrases (such as "This is my password") as still going to be a bad idea. :) So I guess obvious mnemonics are bad, too. - Bilby (talk) 12:16, 14 October 2008 (UTC)

Because long passwords are difficult to remember and they should contain letters (both upper case and lower case), numbers, and symbols, most security systems fail miserably once users get in the habit of writing their password on a post-it note and putting under their keyboard, in their drawer, or even on their monitor. Many of the people I work with are not native English speakers, I've found a nice trick. I taught them l33t. It is easy to remember for most people. Then, I tell them to use a slang word in their native language and type in sort of in a l33t way. So, something like "huozipichu" is easily remembered as "Hu0z1P1c#u". I don't know much more than English, so I decided to use slang words I learn from others. My problem is that if I don't use a password for a while, I forget the slang word and then can't remember the password. -- kainaw™ 12:59, 14 October 2008 (UTC)

Another approach for passwords you don't enter often is to take a random sentence from a book that you have near the computer. Just remember what page it is on. A passphrase is much more secure than a password, as long as nobody knows what book you used. If you type it often enough it becomes quite easy to remember, incidentally, because it is not just random letters and characters. --140.247.11.9 (talk) 23:13, 14 October 2008 (UTC)

Extracting *.rar files in Mac OSX

I have an archive that I need to extract. It has lots of files ending in ".rar", and before the file extension, they are all numbered - like archive1.rar, archive2.rar, and so on. I've tried using The Unarchiver and Stuffit Expander, but neither of them work. Stuffit says "An error occurred attempting to extract... The structure of the archive is damaged (Error #17540)." I hit OK, and it starts on the next file. It continues on to the next file, and keeps on giving me the same error. I'm pretty sure it's not corrupted, because I extracted it once before.

Any suggestions?My name is anetta (talk) 12:20, 14 October 2008 (UTC)

Use Puppy Linux for Mac. You can start it from a pen drive. Mr.K. (talk) 12:37, 14 October 2008 (UTC)

I'll (try to) run it in VMWare Fusion, and presume it can extract *.rar. From your answer, I'm not sure if you noticed, so I'll make it explicit: this question is about extracting an apparently corrupt archive. If using Linux will help, I'm willing to try. But running Linux isn't the aim, unlike my question further up the page about using Ubuntu in VMware Fusion.My name is anetta (talk) 13:00, 14 October 2008 (UTC)

I've never tried it, but assuming that the problem is that the unarchivers you've tried simply failed to handle multi-part rar archives, (rather than a corrupted file), you might want to give UnRarX a shot. - Bilby (talk) 13:05, 14 October 2008 (UTC)

Yes, I am aware that your files are probably not corrupted. Having extracted them once, however, doesn't mean that they are not corrupted now. Perhaps the easier way would be to use a computer with Windows or Linux on it. 80.58.205.37 (talk) 15:09, 14 October 2008 (UTC)

You can always try the official UnRAR command-line program; scroll down to "UnRAR for Mac OS X - Command line freeware Mac OS X universal binary UnRAR" --71.106.183.17 (talk) 17:48, 14 October 2008 (UTC)

Thankyou, 80.58, I presume you're Mr.K logged out. The file is extracted anyway, thanks again.My name is anetta (talk) 17:49, 14 October 2008 (UTC)

I'm a Windows XP user, but Zipeg is some awesome freeware that is also compatible with Mac OSX. It can open pretty much any archive file: ZIP, RAR, ARJ, LHA/LZH, 7z, TAR, GZ, TGZ, BZIP2, CPIO, RPM, ISO, CHM, Z, CBR, CBZ, WAR, and EAR, to name some. I've used it for quite some time as my default extractor, and it works fine with RARs, from my experience. bibliomaniac15 03:16, 16 October 2008 (UTC)

directional ethernet cable?

after a couple of unsatisfactory days trying to get IT at work to determine why i couldn't log in to the network, i finally got some guy who explained to me that the ethernet cable we use has a little H at one end for the hub end and a little S at the other end for the system and mine might be reversed; and in fact, it was and hooking the H end up to the little connector on the wall and the S end up to the computer fixed everything. ?? The ends look quite symmetrical; it's a crossover cable, but that's still symmetrical. the cable says 568-A Cat 5. i screwed around with serial cabling and breakout boxes and null modems and all that crap for a long time,but i've never heard of ethernet cable being directional. how's it work? how common is that? my home router manual doesn't say anything about it, although it does address the crossover/no crossover issue. what's the deal? thanks. Gzuckier (talk) 16:33, 14 October 2008 (UTC)

These guys say if you're connecting a PC to a hub "you MUST use a Straight cable." Looking at their page for a straight cable, I can't see any way that swapping ends would make a difference. I'm puzzled too. --LarryMac | Talk 19:52, 14 October 2008 (UTC)

My first assumption on reading the original question is that it's a little gentle trolling. But I'll assume good faith...

Ethernet cables are not directional. No ifs, no buts, the concept makes no sense. Whoever told you that was either messing with you, or had a valid motive but a questionable way of implementing it. The rationale is that if I as Helldesk person ask one of my users to please physically check that a given cable is plugged in, and in the correct socket, there's a fair chance that he'll just say "OK", sit there for a few seconds, and then tell me he's checked and all is fine. In his mind this is not the problem, he's sure it's plugged in, and he's not going to go crawling around under the desk to check what he already knows to be true. In a bid to make sure he actually does go look at the connections, I can ask him (on whatever pretext - reversing cable, blow into plug, etc) to remove and replace the cable. If he tells me he's done that (and he didn't see through my request for the sham that it is) I can be much more confident that he actually went and looked at some cables.

It's a well-known ploy. 81.187.153.189 (talk) 19:59, 14 October 2008 (UTC)

I'd considered that the IT guy was fooling around, but what about the alleged H and S labels? --LarryMac | Talk 20:07, 14 October 2008 (UTC)

The 569-A refers to the standard now known as TIA/EIA-568-B. The ends are interchangeable. --—— Gadget850 (Ed) ^talk - 20:20, 14 October 2008 (UTC)

me again. serious, not trolling. if i can, i'll upload photos of the labels at the ends, but that doesn't really prove anything... anyway, being careful about what end of the cab le goes where seems to make the absolute difference, just as the guy said. all the references from all the related articles here don't mention anything, every one says direction doesn't matter, and the other thought i had, i.e. only one end is grounded, is also defined as untrue.

idea, however: termination? wouldn't it be bad for the network to have unterminated cables sticking out of the hub? i can imagine that the connector in the box on the wall may have a little switch inside that puts a termination across it when the cable is removed, and the same for a laptop docking station, but a lot of the cubicles here have just the cable hanging out of the wall box without anything on the computer end when somebody takes the computer away... so is it possible that the cable could contain a termination on one end that was sufficient when there wasn't a computer plugged in, but wouldn't "short out" the signal to a computer that was plugged in? of course, that wouldn't exactly explain why it would "short out" the signal if the wrong end was plugged into the wall..... just grasping at straws..... Gzuckier (talk) 15:39, 15 October 2008 (UTC)

The termination theory doesn't really make electrical sense. By not being plugged in to a device on both ends, the cable in essence is terminated. No data can flow through an unplugged cable (on either end), as there is no complete circuit. Is it possible that your company doesn't use standard Ethernet cables and network equipment for some reason? Because if a cable was made by your company and not wired (in the little plug housing thing) in the standard order, then direction would make a difference. Say the 8 conductors are numbered 1-8, if end A was arranged 12345678 and B was 87654321 or if B was also 12345678 (pass-through and cross over, although I am just making up arrangements) then it wouldn't matter which end was where. But if either end was scrambled, say A was 12345678 and B was 3657814 (again, just making up gibberish to demonstrate my point), then the correct plug on the correct end would make a difference. But I can't think of any reason to do that, so that's probably no more than a theory... (see Ethernet crossover cable#Crossover cable pinouts if you want the actual arrangements, there's a nice diagram) DaRkAgE7_[Talk] 03:11, 16 October 2008 (UTC)

Well, there's a quick test to determine if your cable is directional. Unplug both ends (again) and re-reverse it to its original state, making sure both ends are plugged in snuggly. Matt Deres (talk) 16:25, 15 October 2008 (UTC)

I just wanted to add how you can SEE how the cable is reversable.

A Picture is worth a thousand words. $--~~~~$

SMS services that support Primus Canada?

Are there any free services that provide weather updates, or alerts from arbitrary RSS feeds, via SMS and support Primus Canada? (Alerts via e-mail won't help, because Primus doesn't provide an e-mail address for my phone.) NeonMerlin 20:45, 14 October 2008 (UTC)

What would be the best WiFi technology for a very busy and large house/office?

We live in a big Penthouse and use it as an office too. The house is very big and it has 2 levels. Also, we're located at the 8th floor, with 7 floors of offices bellow, which have their own WiFi network. We use wireless phones, which operate on the 2.4 Ghz freq.

I've been reading a lot about 802.11n and a bit about 802.11y. I would like to know if any of these, or any other, technologies would be better for our penthouse.

I must say that it has all kinds of halls, doorways and windows. All made of drywall and some real concrete walls too. Mainly it's a square. Renich (talk) 22:59, 14 October 2008 (UTC)

THIS is a tricky question, but MY suggestion is get a router and a wired network. if you use it for any type of office work just the security alone would be a large benefit. Assuming though you don't want that then that leaves other options. 802.11n is not officially out yet, so any router that claims it is, means it is "pre-n" and for all we know n could change drastically from the state its in now, leaving proprietary issues. Not my suggestion. Likewise, 802.11y isn't out yet, it's just a theoretical band with no actual routers supporting it. So unless this is in the future you need this set up, I can't suggest that. That leaves you down with 802.11a/b/g, and G is a combination of a and b more or less. I would suggest g if you HAVE to use wi-fi. Now for the channel. Basically use whatever you can that's not near other channels. use whatever the floors below you arn't using, and separate yourself from the 2.4Ghz freq. That's the most I can suggest offhand. The problem is the network won't penetrate the concrete walls very well, but the drywall it shouldn't have significant problems with.

In short, Wire your network, If you can't then get a 802.11g router, but if you can, wire it please. cat5e-cat6 doesn't cost THAT much. Forai (talk) 23:07, 14 October 2008 (UTC)

Thanks for the quick reply. The problem is that most of the computers here are laptops and that leaves me with setting up 4 routers and putting them around the house, bridged. And, isn't the unfinished standard correctable by a firmware update in the future? Renich (talk) 23:17, 14 October 2008 (UTC)

The unfinished standard MAY be correctible, what pre-n does is it takes the basic premise of what 802.11n is going to be about (dual antenna, bouncing as I understand) and it creates a router that uses that. For all they know they could say "802.11n is HORRIBLE" and turn the standard into an advanced way of creating a unicorn. OBVIOUSLY a stretch of an example but they could say "research has found three antenna are SIGNIFICANTLY better at reducing noise, and increases distance threefold" or something like that, and impliment 3 antenna. BASICALLY I just don't trust a company coming out with a product before the standards say it should. I get what you mean about the laptops being drastically far apart though. My suggestion in that case is 802.11g for wifi, but the point is still wired IF PLAUSIBLE or possible.

in short, 802.11g. Wired is still the PLEASE DO suggestion but wiring does require alot of work. —Preceding unsigned comment added by Forai (talk • contribs) 23:34, 14 October 2008 (UTC)

Assuming your laptops have support for 802.11n(draft), I'd suggest going with that. I frequently connect to wireless N networks, to great success. While the standard may not be completely finalized, it's close enough that I don't think it will change much. If they do change it, any router you buy now will continue to work in the future. (Honestly, I thought N had been finalized by now.) Anyway, I love the speed and range of N, and I always get really strong signals, even through multiple walls. Of course, make sure to secure your network immediately (not only to protect your files, but to keep WiFi moochers like myself from using Limewire on your connection), using something like WPA (of course, don't use "password" as your password... how many free internet sessions that one has gotten me). Anyway, yeah, I think 802.11n is definitely the way to go... I don't understand why it's not more widespread already. DaRkAgE7_[Talk] 02:41, 15 October 2008 (UTC)

After setting up hundreds of networks over the years, and currently using WiFi. The best answer is NONE. ( well, mabye the presidents office, and the largest confrence room ), but other than that, just run eveything along baseboards. I wired one design firm that *hated* blue wires. The DETESTED them, so we encased all the blue wires in 1/2 PVC that they painted with a gold stripe. Looked great, and every workstation had 100base speeds. ( They did their confrence room in wireless, a linksys, that had multiple problems, WEP got cracked, it had multiple firmware issues... replaced it with a NetGear, and they were happy as clams. Location was SOM in San Francisco ). ( Oh, did I tell you I am hijacking someone else's network? ) Secret (talk • contribs)

Is there a relative to Flash in GNU & Linux?

I've been looking for a while and I can't find a tool or language that does the same things as Flash does. Don't get me wrong, I'm not interested in Flash at all, but, maybe, in the functionality it offers, if it's available for GNU & Linux. I'm looking for some kind of SVG framework or something that enables me to make interactive presentations fast and easy, as Flash or Director would. Renich (talk) 23:26, 14 October 2008 (UTC)

So, let me get this straight: you're looking for both a different language and an easy tool and you want it to run inside Linux? No, of course not. Even the open-source community is embracing Flash with the Gnash player. There are programs to make animated SVGs easily in Windows and there are programs to make Flash easily in Linux, but they apparently wouldn't suffice for your needs. I guess you'll have to code an animated SVG in vi. O'Reilly has a book about SVGs that will tell you how.--Account created to post on Reference Desk (talk) 02:21, 15 October 2008 (UTC)

I'm not sure "embraced" is the word you're looking for. I'd go with "is forced to support". APL (talk) 04:01, 15 October 2008 (UTC)

Well, I know that you can view content made in flash, on GNU & Linux, but, what I'm talking about is generating the animated/interactive content. Yeah, I'm sure you can do cool stuff using SVG, but I would like to know if there is some framework or something... or an IDE so I can, or any user, can make, with proper knowledge, a fine and interactive presentation. Renich (talk) 23:02, 16 October 2008 (UTC)

What capabilities are you looking for specifically? I haven't used it, but depending on what you're looking for this might do the trick : [1] APL (talk) 04:01, 15 October 2008 (UTC)

Well, I would say: a) audio/video capable, b) scriptable, c) capable of conections (smtp, ftp, http, etc), d) animanimable, e) easy to run in any environment (with the proper engine/run environment. I want to make cool and interactive presentations on GNU & Linux, if it's with some advanced capability of Open Office, it's ok, just want to know what's in the menu for this. Renich (talk) 23:02, 16 October 2008 (UTC)