Wikipedia:Reference desk/Archives/Computing/2011 November 1

From Wikipedia, the free encyclopedia
Computing desk
< October 31 << Oct | November | Dec >> November 2 >
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.


November 1[edit]

Joining Firefox[edit]

Thank you for this service. At 83 years of age need contemporary help. Recent experience with an American anti-virus software company and their now Indian off-shore technicians, led me to extensive complaints of their delinquent service which has turned out to be a scam, and a costly one at that to many people. This led me to a search of any companies not associated with the Indian complex. My first step was to find if it was possible to operate my computer without Microsoft products!Products that came with the purchase of my computer. As I am totally ignorant of any technical expertise will my downloading of the new Firefox assist me in achieving some better measure of control and security that I feel is not available now due to my recent experience? Hope this provides enough information for your consideration. — Preceding unsigned comment added by 121.222.99.145 (talk) 03:48, 1 November 2011 (UTC)[reply]

Opinions vary as to which browser is the most secure.[1][2][3] Firefox is probably better than Internet Explorer, though Opera and Google Chrome also are good (Opera scores highly, partly because so few people use it), and more recent versions of Internet Explorer have more security features than earlier versions.
More important than the choice of browser is to keep your browser up to date (most now have auto-update features), install anti-virus software, and follow general advice for safe browsing. See for example[4][5][6][7]. --Colapeninsula (talk) 10:44, 1 November 2011 (UTC)[reply]
It sounds like installing anti-virus software was the cause of the poster's trouble in the first place. Install widely recommended anti-virus software, or as Mr.98 says in a thread further up this page, not widely unrecommended software. Making a good choice is somewhat perplexing, even for us contemporaries. Here's the list of antivirus software.  Card Zero  (talk) 13:04, 1 November 2011 (UTC)[reply]
I am still of the opinion that it is safer to run a modern operating system without installing third-party antivirus programs. Proper computer security depends entirely on careful control of what programs you choose to run, and who you give permission to access to your machine. There are very few instances of an operating system security flaw that can infect a machine without user-action (that is: almost all computer infections in recent years occur because the user intentionally chooses to install the virus, usually because they are misled about the malicious intent of a particular program). For example, read about computer security on OS X, on Microsoft Windows; and of course, for those who use Linux, you should know how your system works if you plan to protect it. I have yet to find any convincing evidence that commercial or non-commercial "anti-virus software" serves any purpose whatsoever. If the operating system is uncompromised, no antivirus software is needed. If the operating system is compromised, no antivirus software will help. Nimur (talk) 16:50, 1 November 2011 (UTC)[reply]
This is more or less my philosophy too - I rely on a software firewall and various Sysinternals products to help me look for suspicious behaviour when I get worried, instead of running any intrusive virus checker that's always on. If I do get a virus or persistent malware (once this happened completely spontaneously, from merely visiting a website straight to having bogus software installed and showing up in the system tray) I then research it and find out what painstaking steps are needed to manually remove it (and what I should do to stop it happening again). I'm not sure about your last point, though, since eliminating a virus by hand can often be a lot of work, and antivirus software ought to do the job automatically (if it knows the particular virus). Probably it would be more sensible (but less interesting) to use some.  Card Zero  (talk) 14:26, 2 November 2011 (UTC)[reply]
Note that Firefox is not specifically security software - it is a web browser that allows you to view websites such as this one, instead of using Internet Explorer. Browsers can be a source of security issues, but is difficult to make a definitive decision about which is most secure - people often use alternative browsers because they have more features or are easier to use, not necessarily because of security issues. Since it sounds like you are using Windows, Microsoft (the company that sells Windows) maintains lengthy lists of 'partner' antivirus software providers for recent versions of Windows at [8] - presumably you can safely assume that they are legitimate (and should work with your computer). Some of these antivirus software can be used for free, though they may occasionally nag you to buy some kind of 'premium' version of the software. Other operating system providers should have security information on their sites too - GNU/Linux and Mac OS X are often held to be more secure than Windows - partly because they are less widely used and hence less likely to be targeted by malware. Government and police websites can also be a useful and trustworthy source of computer security information. For example, the UK government maintains Get Safe Online (which is recommended on the official government website here). 130.88.73.65 (talk) 11:57, 3 November 2011 (UTC)[reply]
Actually, since it appears you are in Australia (sorry if this is incorrect), the Australian government's equivalent seems to be Stay Smart Online 130.88.73.65 (talk) 11:59, 3 November 2011 (UTC)[reply]

Questionable ways of improving code[edit]

I once read about an American tech company that challenged each job candidate to optimize certain blocks of code. Unbeknownst to the candidate, the code segments were speed bottlenecks in the production code, and the senior developers couldn't figure out how to optimize them further. To the humiliation of the developers, a recent college grad discovered a linear algorithm for one code segment (that was initially exponential); a high school girl discovered a one-line mathematical solution to another. The developers proceeded to use these 2 solutions without hiring the applicants.

My questions are: 1. Does anyone recognize this story? I read about it years ago, and have since forgotten where I read it or what the company was 2. Is this clever method of getting other people to improve your code legal? --140.180.14.123 (talk) 06:29, 1 November 2011 (UTC)[reply]

We can't offer legal advice, so what follows are general comments. Computer software is covered by copyright law in most jurisdictions. However, this doesn't protect algorithms (patents can protect algorithms, at least in some jurisdictions, but must be applied for - see here for the distinction). So reusing a general idea would probably not be illegal, particularly if the method is already common knowledge, but directly using code written by a job applicant might be breach of copyright. If the company had little or no intention of offering a job, the company might be guilty of fraud, obtaining services by deception, misleading or deceptive conduct, etc, or liable in civil law for claims for fraud. --Colapeninsula (talk) 11:13, 1 November 2011 (UTC)[reply]
The story sounds a bit like an urban myth to me (though that's not to say it hasn't ever happened), similar to the Unsolvable Math Problem. AndrewWTaylor (talk) 11:58, 1 November 2011 (UTC)[reply]
(I just asked at Wikipedia:Reference desk/Science#George Dantzig what the two problems were.) Comet Tuttle (talk) 01:19, 2 November 2011 (UTC)[reply]
It sounds unlikely to me. A development shop that was incompetent enough to use an exponential-time algorithm when a linear-time one was available would probably also have trouble (a) identifying troublesome pieces of code, and (b) attracting competent interviewees. Paul (Stansifer) 12:22, 1 November 2011 (UTC)[reply]
Contrast it with this: Every IT company I've worked for has had a bit in the employment contract about any 'inventions' remaining the property of the company. For example, if I, an employee of the company, was to invent a new image compression algorithm, my contract would prevent me from patenting it myself and instead my employer would be the ones legally able to exploit that invention. I would be surprised if the same didn't apply to interview candidates, though the company might want to be sure of this by asking candidates to first sign a Proprietary information agreement of some kind before the (bogus) test. Whether this has actually happened, I am unsure, though the thought did occur to me during the last programming proficiency test that I took. Astronaut (talk) 12:31, 1 November 2011 (UTC)[reply]
As far as I know, in my company, this only applies for inventions made during work time. If I invent a new compression algorithm when sitting at the company's office, and getting paid for my time, the algorithm becomes the company's property. If I do it when sitting at my home, not getting paid for my time, it remains my property. 194.100.223.164 (talk) 11:11, 2 November 2011 (UTC)[reply]
Thanks for the answers. I don't know about a), but I don't think b) was a problem: the interviewees were not competent. The recent grad probably had little to no working experience; the high school student was probably just very clever and knew a lot of math. --140.180.14.123 (talk) 17:08, 1 November 2011 (UTC)[reply]

Adobe Photoshop CSS 5.1 crashes with every attempt to create GIF[edit]

Just when I'm right at the tail end of finishing the process of making a GIF (from a video), I try to save my work as a GIF, and that's right when it decides to freeze and force close. EVERY. TIME. And I always lose my work, forcing me to restart from scratch. What the hell is going and what can I do to rectify this? 174.95.227.232 (talk) 07:26, 1 November 2011 (UTC)[reply]

Do you get a message with any details? Was this always the case, or did it only started behaving like this recently? At what point exactly does the freeze occur? --Ouro (blah blah) 10:41, 1 November 2011 (UTC)[reply]
All I get is a "Windows is searching for a solution to this problem" dialog box, and the "solution" is a forced shut down of the program. It freezes right as I'm done with the editing process and am about to preview it. I only just begun using it (I have to add that it's a free trial version), but it's happened about 3-4 times in a row. 174.95.227.232 (talk) 11:00, 1 November 2011 (UTC)[reply]
Is it a new system? Do you have problems with other functions of the software? Do you get other video playback issues? --Ouro (blah blah) 15:17, 1 November 2011 (UTC)[reply]
I'm using Windows 7, and I haven't really used the other functions of the software, so I don't know. 174.95.227.232 (talk) 00:49, 2 November 2011 (UTC)[reply]
I don't know a solution, but the first piece of advice is to save your layered image as a PSD before trying to turn it into a GIF, so that at least you won't lose all your work. Looie496 (talk) 17:45, 1 November 2011 (UTC)[reply]
I was going to suggest this too - I may be about ten years out of date, but I was thinking in terms of Photoshop loading a plugin (the Quicktime plugin) for the codec when it wants to extract frames from a movie, which might then make Photoshop unstable until it's restarted: so putting a restart of Photoshop between extracting the frames and saving the GIF might help.  Card Zero  (talk) 14:34, 2 November 2011 (UTC)[reply]

If you can save it as something else, we can explain how to get a GIF from that using other software. ¦ Reisio (talk) 17:49, 1 November 2011 (UTC)[reply]

Please do, I'd gladly appreciate it. 174.95.227.232 (talk) 00:49, 2 November 2011 (UTC)[reply]
I wonder if VirtualDub would be enough for the OP. Apart from the fact that I haven't got a clue as to what could be wrong with his software/box. --Ouro (blah blah) 06:12, 2 November 2011 (UTC)[reply]
Are you doing Save for web? If so, try holding down ctrl + alt, while at the same time clicking (without releasing) the file menu -> save for web. A dialogue box will appear and ask if you want to reset your preferences. Say yes. TheGrimme (talk) 18:13, 1 November 2011 (UTC)[reply]
Didn't make a difference, I'm afraid. 174.95.227.232 (talk) 00:49, 2 November 2011 (UTC)[reply]

VBA for Microsoft Word help[edit]

Dim x as Integer

x = Documents.Count

MsgBox x

I have 14 documents open, why does the message box say "1"? — Preceding unsigned comment added by 203.116.187.1 (talk) 08:59, 1 November 2011 (UTC)[reply]

I would suspect that you actually have 14 instances of Word running, each with one document open. Check the Processes tab in Windows Task Manager to see if there WINWORD.EXE occurs more than once in the list. How are you opening the documents? It might also be helpful if you could tell which versions of Word and Windows you are running. AndrewWTaylor (talk) 11:55, 1 November 2011 (UTC)[reply]

Yes, I have 14 instances of Word running. I opened the documents by clicking a button in a program of my company's intranet. I am using Word 2003 on Windows XP. How can I make Documents.Count count the documents in all processes or make Word open all new documents in the same instance of Word? Thanks for the help. — Preceding unsigned comment added by 203.116.187.1 (talk) 03:41, 2 November 2011 (UTC)[reply]

regex[edit]

I'm using a Windows version of grep. I want to match url links from a html file and display them. I tried

grep.exe -i http://(.*)\.(.*)/(.*) file.html

which sort of works but lists </a> tags and other stuff after the url. How do I make it stop before </a> tags or " characters? 82.43.90.142 (talk) 11:29, 1 November 2011 (UTC)[reply]

The final .* matches everything. Even the first .* matches everything. What you are asking for is http:// followed by as much as you can find until you see a period. Then, as much as you can find until you see a /. Then everything else you can find. You need to add some limitations on what you accept. I don't use Windows grep, but it looks like the regular grep syntax is the same on Windows as everywhere else. So, you use [blahblahblah] to limit what you accept. For a URL domain name, the valid characters are a-z, A-Z, 0-9, -, _, and . (I think that is all of them). So, you match http://[a-zA-Z0-9-_]*/ to match the domain name. Now, you want to continue matching, but you want to stop if you see something that shouldn't be in the URL. To say "I don't want", you add a ^ inside the square brackets. Let's assume that a ' or " follows the URL at all times. You say you want everything EXCEPT those characters as in http://[a-zA-Z0-9-_]*/[^\'\"]*. I assume you need to escape the quotes. But, you can see that the domain name is everything except the forward slash. So, we can simplify to http://[^/]*/[^\'\"]*. Then, just to add another way, you can use the -P option to turn off greediness. Then, grep won't go looking for as much as possible. -- kainaw 13:59, 1 November 2011 (UTC)[reply]

Thanks 82.43.90.142 (talk) 15:03, 1 November 2011 (UTC)[reply]

Or just use the quantifier *? instead of * See below
The second one grabs as many characters as it can, so it is said to be greedy. The first one does the opposite thing, so it is said to be reluctant. Therefore, (.*?)\. grabs the minimum amount of characters which is followed by a dot.
@Kainaw, instead of listing the possible characters, one could just write [^.]
Source: [9] --151.75.36.35 (talk) 17:08, 2 November 2011 (UTC)[reply]
I don't think the gnu version of grep supports non-greedy matching operators like *?. If it does your version might be an extended set of tokens? Broba (talk) 22:26, 2 November 2011 (UTC)[reply]
Oh, yeah, it seems that gnu grep uses a text-directed engine, which does not support lazy (i.e. reluctant) quantifiers. [10] I didn't know that, I'm sorry. --151.75.107.247 (talk) 00:37, 3 November 2011 (UTC)[reply]

unable to chat - where do I type? (webchat.freenode.net)[edit]

In the webchat it says: "Just ask your question, and one of our friendly helpers will answer as soon as possible :-)" But where can I type my question? There is no dedicated field... I've tried just typing, but nothing happens. Help! Feeling terribly stupid since I'm unable to do the most basic thing. In case it matters/might be the source of my trouble?: I'm using Mac OS 15.8 + safari - Thanks! Lilly333 (talk) 13:51, 1 November 2011 (UTC)[reply]

I don't see anything about 'just ask your question'... is that perhaps on a website directing you to freenode.net? I have no problems connecting to that site using the most recent MacOS/Firefox combo, but if you really want to connect to this IRC server and your webbrowser won't let you, perhaps you can try a dedicated IRC client? Internet_Relay_Chat#Clients Unilynx (talk) 17:47, 1 November 2011 (UTC)[reply]

At the very bottom of the viewport (just above the status bar in browsers that still have them), there will be a white/ish area where you input your messages, and hit ENTER to send them. http://webchat.freenode.net/?nick=Lilly333&channels=#MacOSX ¦ Reisio (talk) 17:53, 1 November 2011 (UTC)[reply]

It worked for me, too, using Windows 7 and Firefox. Comet Tuttle (talk) 21:08, 1 November 2011 (UTC)[reply]

Aaaahhhh! Found it! Thanks. Feel really silly now, but I looked everywhere (well actually, not)... thanks again Lilly333 (talk) 11:48, 2 November 2011 (UTC)[reply]

It not your fault, the software should have changed the keyboard focus to that location, so you could just start typing. StuRat (talk) 20:53, 4 November 2011 (UTC)[reply]

internet homepages[edit]

So, running windows 7 and IE9, I recently downloaded something I perhaps should not have, which has changed my homepage. I found out how to change it back, but now every time I open a new tab, I get that homepage again, rather than the new tabs page. So, firstly is there any way of getting the new tabs page back? Secondly, could I instead set it so that when I open a new tab, a different site comes up than my usual homepage, and thirdly, how do I stop this happening again?

148.197.80.214 (talk) 18:13, 1 November 2011 (UTC)[reply]

Take a look at this article, specifically at the Change the New Tab Page section. TheGrimme (talk) 18:16, 1 November 2011 (UTC)[reply]

That answers question 1. And I think shows that the answer to question 2 is no. And, most likely the only real answer to question 3 would be not to download anything again. Though, oddly, links now open in new windows rather than tabs, even after I specifically set for them to be tabs in those same options, yet another problem to try to fix... 148.197.80.214 (talk) 23:53, 1 November 2011 (UTC)[reply]