Wikipedia:Reference desk/Archives/Computing/2016 February 29

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 29

Copypasteable fiction sites

What are some websites where you can go to find a lot of user-submitted fiction to copy and paste? I want to make a corpus of contemporary fiction in English (like this). Fictionpress and WattPad have locked the ability to highlight and copy text. 2601:644:101:84B2:A084:B6B6:3372:7250 (talk) 07:38, 29 February 2016 (UTC)

It is a semantic argument, but a website owner/designer cannot block the ability to copy and paste. They can only request that the web browser does that (usually with JavaScript). It is the web browser that blocks the ability to copy and paste. Once you understand that, it is trivial to switch web browsers or add a plugin (noscript) to your existing web browser to keep it from blocking your ability to copy and paste. 209.149.114.211 (talk) 12:40, 29 February 2016 (UTC)

NoScript is a Firefox extension that allows you to block Javascript. If you turn NoScript on and you visit those websites with Firefox then you'll be able to copy the text. You can download it at http://noscript.net.The Quixotic Potato (talk) 19:26, 29 February 2016 (UTC)

@The Quixotic Potato: I use NoScript all the time (I like my internet broken and clunky ;) and I still can't copy text from FictionPress [1]. It's possible I've just failed to set NoScript to truly block everything it can, or perhaps they can prevent easy copy/paste without using JS? Clarification on this specific site and tool combo might help OP. SemanticMantis (talk) 19:49, 29 February 2016 (UTC)

@SemanticMantis: Weird. It works for me, and it stops working when I temporarily disable NoScript. Of course it is possible to do something similar without using JavaScript (sticking a transparant image on top of the text with CSS would work for example). I've been thinking about turning one of my old computers into a webdevelopment testing machine that will run browsers like IE6 and Firefox without any modifications/plugins, to see the web as others see it, but at the moment I don't have a browser that uses default settings. The Quixotic Potato (talk) 20:25, 29 February 2016 (UTC)

Thanks, looks like whatever problem is only on my end then. SemanticMantis (talk) 21:23, 29 February 2016 (UTC)

I have the problem here too (Chrome, FF and IE11): the text uses the non-standard CSS property user-select, which may not work in all browsers, but can stop you selecting text. You can temporarily disable it by entering javascript:$(".nocopy").css ({"-webkit-user-select":"all","-moz-user-select":"all","-ms-user-select":"all","user-select":"all"}) in the address bar of the browser. AndrewWTaylor (talk) 21:55, 29 February 2016 (UTC)

Archive of Our Own has all sorts of Fanfic. Literotica does too, but it's almost exclusively erotic literature. Fictionhub [2] is more general. This [3] is an implementation of the library of babel, and has every page that can ever be written. Project Gutenberg [4] hosts many many fiction books that are public domain and can be freely copied and reused. Fair warning our article List of fictional books is about books that don't exist, not about books that contain fiction. But I for one would love to see frequency analysis for fictional books :) SemanticMantis (talk) 19:43, 29 February 2016 (UTC)

How do you deal with possible false positives in virus scanning?

When you try to confirm that software you downloaded from the Internet is not infected using a site like VirusTotal, you will often find that a very large majority of the virus engines will say the binary is clean, but a small number of them will indicate that the binary is infected with the same or different malware. How do you tell whether the small numbers of positive results are indeed false positives? --98.115.39.92 (talk) 13:19, 29 February 2016 (UTC)

Well, it is possible to decompile software, but if it is software I do not really care about then I will simply run it in a sandboxed enviroment, like a virtual computer (while logging everything it does with something like Process Monitor). See Sandboxie or VirtualPC. In many cases it isn't very difficult to discover why something is a false positive, you can tell by the name the AV uses, for example because it uses a packer that is commonly used by malware authors and can't be unpacked by the AV. The article Executable compression says: "some older virus scanners simply report all compressed executables as viruses because the decompressor stubs share some characteristics with those. Most modern virus scanners can unpack several different executable compression layers to check the actual executable inside, but some popular anti-virus and anti-malware scanners have had troubles with false positive alarms on compressed executables.". If you have the virusscanner that shows the false positive then its quite easy to discover what the actual detection string is (Google "offset locator"). False positives based on heuristics are a bit more complicated. The Quixotic Potato (talk) 17:29, 29 February 2016 (UTC)

Note also in some cases it isn't so much a false positive as a positive you may not agree with. For example some AV will detect keygens and cracks. Some will detect programs intended for legitimate purposes (e.g. miners, remote access tools) which are often installed secretly. Again the description should make this clear, e.g. it may call it a potentially unwanted program and/or a hack tool. The reason the other AV doesn't list this program may be because of a difference of philosphy or even simply because they haven't encountered it enough and it's probably not a high priority to add as it's not clear malware. Nil Einne (talk) 18:18, 29 February 2016 (UTC)

I installed a legal copy of a former virus protection software which has formatted my boot sector. As all data was stored on anohter machine, the damage was not relevant, but I decided no longer use software from this manufacturer. Some issues come by getting the early update of signatures. Since I set to update all 4 to 8 hours only and update the clients once a day, all problems have gone. Users are instructed not to open sucspicous mail. Sometimes it is an good advice to reinstall the recent antivirus software version on the machine. In case of trouble or slow machine cut the it from the network, remove the virus protection software, clean up files and registry by tool from the software vendor and reinstall the recent antivirus software version offline, downloaded from a machine You trust. In case of doubt, boot the malfunctioning machine from a live-cd, the antivirus software vendor offers, update the signatures online and scan the machine. There are also reliable free anitvirus software versions for noncommercial use avail. --Hans Haase (有问题吗) 12:21, 3 March 2016 (UTC)