Wikipedia:Reference desk/Archives/Computing/2013 January 15
Computing desk | ||
---|---|---|
< January 14 | << Dec | January | Feb >> | January 16 > |
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. |
January 15
editTrying to understand nofollow
editI always thought nofollow meant that a page we linked didn't have its Google search results modified because of our link, and nofollow seems to confirm that. This confuses me substantially, because some days ago, I found an obscure PDF source while searching Google for <"barker village site">, but only roundabout: another document on the same site mentioned it, and by navigating around the website, I found the document in question, so I added an external link to it when I cited it. Earlier today, when searching for the same term, this PDF is the top result on Google. This is an obscure archaeological site; it's not been in the news lately, and the host website (a public library) isn't spamming itself everywhere this week, so I can't see why there's such a big change when the only apparent change is the addition of a link from the nofollow-equipped English Wikipedia. I'm working on a very bad connection right now, and I can't even access Google (not sure how I got this far!) to find the URL, so no link yet. I'll come back and provide a link when I'm on a better connection, as well as fixing any inaccuracies from my memory of earlier. Nyttend (talk) 00:18, 15 January 2013 (UTC)
- Link is here. Nyttend (talk) 01:21, 15 January 2013 (UTC)
- The "nofollow" convention is an unofficial markup tag that Google or anyone else may interpret in any way they choose. There's absolutely no reason to assume that Google will interpret a "nofollow" tag at all; or that they will interpret it in a way that makes sense to you. In fact, Google's official documentation is ambiguous. You should not assume that Google discloses its exact algorithm or methodology - even when it releases documentation or general guidance. The search-engine's algorithm, and its data repository, are both proprietary; Google does not explain why it delivered a particular search-result for a specific query. Nimur (talk) 03:24, 15 January 2013 (UTC)
- I'm confused largely because my results don't seem to follow the cited information in the article's "Interpretation by the individual search engines" section. I understand that Google doesn't publish everything, but I don't understand why this is apparently different from results in other studies. Nyttend (talk) 07:30, 15 January 2013 (UTC)
- I think google adheres to nofollow but there may be some other site that monitor Wikipedia for new references to pdf documents. It would be interesting to know what sites do follow changes in Wikipedia and what exactly is involved in something like that. Dmcq (talk) 12:14, 15 January 2013 (UTC)
- There are also many Wikipedia mirrors, which may declare nofollow, but are not obliged to. One might try to search for some specific text which surrounds the link in Wikipedia and check whether it appears on other WWW sites... If so, it might explain the document's high score in Google results. --CiaPan (talk) 14:28, 15 January 2013 (UTC)
- The "nofollow" convention is an unofficial markup tag that Google or anyone else may interpret in any way they choose. There's absolutely no reason to assume that Google will interpret a "nofollow" tag at all; or that they will interpret it in a way that makes sense to you. In fact, Google's official documentation is ambiguous. You should not assume that Google discloses its exact algorithm or methodology - even when it releases documentation or general guidance. The search-engine's algorithm, and its data repository, are both proprietary; Google does not explain why it delivered a particular search-result for a specific query. Nimur (talk) 03:24, 15 January 2013 (UTC)
Substitution-only Tcl expressions
editWhat is the canonical way to write
proc just x {set x} ;# or "return $x"
foo [if {...} {just $x[bar]} {just $y[baz]}]
preferably without writing my own proc just
? The idea is that I want the value of the if to be one of those two "Tcl words", with their appropriate substitutions, but there's no command I want executed on them directly and I don't want a temporary variable just so I can then get its value with set. (This is a common construct in Lisp, of course: (foo (if (bar) baz quux))
, but there there is no "substitution" as distinct from evaluation. If the expressions were mathematical, I would use expr ... ? a : b, but they're not and they might contain characters that expr treats as operators.) --Tardis (talk) 01:06, 15 January 2013 (UTC)
Laptop inserting extra returns in Mediawiki software
editHi. When I edit with this laptop, and try creating a gap between paragraphs, the software interprets a single return as two.
The lines above and below are good examples, as are my last edits on Ten's user talk.
While BOGOF is great (sometimes) in shops, here it's not so good. Any advice? --Dweller (talk) 12:49, 15 January 2013 (UTC)
- Sure it is not a sticky enter key? Astronaut (talk) 14:05, 15 January 2013 (UTC)
- Hmmm. OK, so it's not a sticky enter key, because it shows on this edit screen as one gap. However, I've just discovered it's having the same effect in Word, although there I see the huge gap. So, it's not just Mediawiki. --Dweller (talk) 14:40, 15 January 2013 (UTC)
- This is an off-the-wall suggestion, but check your language settings - some ancient part of my memory recalls having a similar problem when I set the wrong keyboard language settings. Something to do with interpreting Carriage return/Line feed. 80.195.151.245 (talk) 15:47, 15 January 2013 (UTC)
- Hmmm. OK, so it's not a sticky enter key, because it shows on this edit screen as one gap. However, I've just discovered it's having the same effect in Word, although there I see the huge gap. So, it's not just Mediawiki. --Dweller (talk) 14:40, 15 January 2013 (UTC)
Java problems
editIf my computer was infected through the recently reported Java vulnerabilities, is the problem solved simply by disabling Java, or does the infection/problem remain once the computer is infected, even if I have now disabled Java? What is the best way to check the computer specifically for problems that have been described as related to the Java problems? Will free versions of Avast or McAfee adequately check the computer? Thanks if you can advise. — Preceding unsigned comment added by 94.208.75.76 (talk) 14:44, 15 January 2013 (UTC)
- The Java vulnerabilities were used in several professional (that is, sold by criminals to other criminals for much money) penetration toolkits. These give their masters the abiity to deliver into your system a range of advanced malware, counter-security software, and remote-control software. If I believed my PC had been pwned by a professional computer crime outfit using one of these kits, such is their sophistication (and the novelty of many of the kit's components), I personally would not have sufficient confidence that any anti-malware solution (paid or not) would be certain to remove everything, and I'd blank the disk and restart from scratch. -- Finlay McWalterჷTalk 15:07, 15 January 2013 (UTC)
- What Finlay said. Backup your personal data to some place you can keep it more or less quarantined, to limit any possible re-exposure to your incoming new system by older files. Reinstall your OS and scan and ferry over the files from your old system that you’re sure to be using. Then make a backup, so you can easily do a full restore to a system that’s ready for you to use. Then set up some kind of intrusion detection system that keeps track of file system integrity, so if this should happen again you won’t have to reinstall to be sure your system is clean. ¦ Reisio (talk) 17:24, 15 January 2013 (UTC)
If I hotlink wikimedia....
edit... Than whose bandwidth is consumed, mine or Wikimedia's.EatIcecream2 (talk) 18:03, 15 January 2013 (UTC)
- Wikimedia's. -- Finlay McWalterჷTalk 18:05, 15 January 2013 (UTC)
- Actually, it's more complicated than that. Network traffic occurs and data has to flow from source (the server) to destination (the viewers). This "uses" bandwidth on both ends. More technically, you'd be instigating data traffic that would occupy bandwidth on every hop from the Wikimedia source server(s) to every individual who views the images. This is further complicated by the numerous caching systems that are in widespread use around the internet. Consider reading RFC 3272, Overview and Principles of Internet Traffic Engineering. Nimur (talk) 02:14, 16 January 2013 (UTC)
- So, do you mind if I hotlink 50 20mb images?EatIcecream2 (talk) 18:15, 15 January 2013 (UTC)
- Those of us with limited internet access would prefer smaller images, or at least a warning if 20Mb is essential. Dbfirs 18:20, 15 January 2013 (UTC)
- It's not up to me; it's Wikimedia's bandwidth, you'd have to ask them. I don't know where you'd ask them (here is not the correct place) - perhaps one of the technical IRC channels -- Finlay McWalterჷTalk 18:22, 15 January 2013 (UTC)
- commons:COM:HOTLINK says "Hotlinking is allowed from Wikimedia servers, but not generally recommended".EatIcecream2 (talk) 18:25, 15 January 2013 (UTC)
- I guess one might want something like that for those panorama pictures or ones where one can zoom into detail though I'd do the same thing as is done with maps of using lower level zoom images and tiling. The business of internet speed is a worry even for serving iPad devices. Dmcq (talk) 15:10, 16 January 2013 (UTC)
- It's not up to me; it's Wikimedia's bandwidth, you'd have to ask them. I don't know where you'd ask them (here is not the correct place) - perhaps one of the technical IRC channels -- Finlay McWalterჷTalk 18:22, 15 January 2013 (UTC)
Where to ask for help in old Commodore 64 games?
editLately I've been playing Jumpman Jr. on VICE on my Fedora 17 Linux system. Levels 1 through 3 were easy enough, level 4 was a bit of a challenge, but I simply can't figure out (pun intended) level 5. Playing on an emulator instead of a real Commodore 64 has the benefit that I can snapshot the game at any moment and come back to it as many times as I want. This allowed me to play through Wizball and Uridium without ever losing a life. However, in Jumpman Jr. level 5, what keeps me puzzled are the bombs that cause the floor beneath them to vanish when I defuse them, thus killing Jumpman instantly. Where can I go to ask for help in old Commodore 64 games? JIP | Talk 19:01, 15 January 2013 (UTC)
- When I was actively using Commodore 64 emulators, I remember Lemon 64 was a neat looking site. It looks like it has an active forum. Maybe they can help or point you to a better place. --Bavi H (talk) 02:33, 16 January 2013 (UTC)
Apple's PAGES app.
editIs there anyway to get a word count on this programme please?85.211.131.113 (talk) 19:04, 15 January 2013 (UTC)
- Yep, for the Mac version anyway, I dont know if you mean the mobile version. For Pages in MacOS (i'm using Pages '09): Go to the Edit menu, select submenu Writing Tools near the bottom, then Show Statistics. By the way, I'm surprised that typing "word count" in the Help menu doesn't bring this up, Apple should fix that. El duderino (abides) 06:04, 16 January 2013 (UTC)
Excellent, many thanks for the help.85.211.131.113 (talk) 07:52, 16 January 2013 (UTC)