Wikipedia:Reference desk/Computing

Welcome to the computing section
of the Wikipedia reference desk.

skip to bottom

Select a section:

• Computing
• Entertainment
• Humanities
• Language
• Mathematics
• Science
• Miscellaneous
• Archives

Want a faster answer?

Main page: Help searching Wikipedia

   

How can I get my question answered?

• Select the section of the desk that best fits the general topic of your question (see the navigation column to the right).
• Post your question to only one section, providing a short header that gives the topic of your question.
• Type '~~~~' (that is, four tilde characters) at the end – this signs and dates your contribution so we know who wrote what and when.
• Don't post personal contact information – it will be removed. Any answers will be provided here.
• Please be as specific as possible, and include all relevant context – the usefulness of answers may depend on the context.
• Note:
  • We don't answer (and may remove) questions that require medical diagnosis or legal advice.
  • We don't answer requests for opinions, predictions or debate.
  • We don't do your homework for you, though we'll help you past the stuck point.
  • We don't conduct original research or provide a free source of ideas, but we'll help you find information you need.

Ready? Ask a new question!

How do I answer a question?

Main page: Wikipedia:Reference desk/Guidelines

• The best answers address the question directly, and back up facts with wikilinks and links to sources. Do not edit others' comments and do not give any medical or legal advice.

See also:

• Teahouse
• Help desk
• Village pump
• Help manual

April 4

What physical principle was used to differentiate a hole from no hole with IBM punched cards?

Was it a conductor and its contact surface the card was dragged between able to conduct on a hole but not able when there was no hole? Or some other principle? 75.75.42.89 (talk) 01:23, 4 April 2014 (UTC)

See Punched card. Various IBM (and other) card readers through the years used mechanical, electromechanical, optical, pure electrical (electrographic), and probably other methods of reading. --Carnildo (talk) 01:37, 4 April 2014 (UTC)

Maybe I'm blind, but I didn't see any even cursory descriptions of the inner workings of card readers of different technologies at Punched card. 75.75.42.89 (talk) 09:10, 4 April 2014 (UTC)

We have an article on Punched card input/output but that is also woefully inadequate.--Shantavira|^{feed me} 11:57, 4 April 2014 (UTC)

A bit in keypunch. I never did much with keypunch, but the missile system I started on in 1978 used punched tape. The tape reader had recently been upgraded from whisker contacts to optical (used the same light bulb as the VW dome light). --  Gadget850 ^talk 12:34, 4 April 2014 (UTC)

And for Tabulating machine. This shows a 19th century method, but thimbles of mercury soon passed out of style. Jim.henderson (talk) 12:31, 4 April 2014 (UTC)

For a electromechanical system, the hole is sensed by an angled conductive brush - if a hole is there, the brush end goes through it and contacts a conductive roller on the other side of the card, completing a circuit; if there's no hole, the card prevents the brush touching the roller. A diagram, for an IBM model 82 card sorter is here. Textbook on Management Information Systems By D P Nagpal (which I found on Google Books) has a schematic diagram for this, and for an optoelectric system, with a bulb on one side and a photocell on the other (fig 3.7 page 66). There's more info about the brush system in this page. A logical diagram of the electrical circuitry behind the brush sensor reader is at this page. Even Herman Hollerith's original systems (patent US395782) were electromechanical, unlike a purely mechanical system like a player piano. -- Finlay McWalterᚠTalk 13:18, 4 April 2014 (UTC)

Our article Plugboard discusses how early card readers worked in some detail.--agr (talk) 19:28, 7 April 2014 (UTC)

Filtering Hindi characters in Unicode

I'm working on Hindi tweets (in Python) and have to scrape away every other language that appears in the each tweet other than Hindi. Is there any way I can specify a range of Unicode values which applies only for Hindi characters so that non-Hindi characters are removed automatically (excluding punctuation marks, which I want to keep)? La Alquimista 06:03, 4 April 2014 (UTC)

You would probably want to use the Devanagari Unicode block which is in the range U+0900 to U+097F. --Canley (talk) 10:53, 4 April 2014 (UTC)

not responding

Firefox (latest version, 28.0) freezes 3-4 times a day, for about 10 seconds, with the message "not responding." Adobe Reader (v. 10.1.9) has also done so, and I believe other applications as well. What gives? Anything I can do about it? I have an HP 2000 laptop running Windows 7. --Halcatalyst (talk) 13:50, 4 April 2014 (UTC)

Similar problems on my (older) laptop are caused by an anti-virus scan running to schedule, and also occasionally by my lack of memory (with lots of pages being switched in and out of pagefile.sys) You might like to run performance monitor (and resource monitor) to see if any programme is hogging resources. Dbfirs 08:12, 7 April 2014 (UTC)

How to recover my bookmarks after a Windows 8 refresh?

I have used Windows 8’s “Refresh your PC without affecting your files” option, and I have lost in the process all my Firefox and Chrome bookmarks. No restore point had been created prior. — Preceding unsigned comment added by 69.123.17.140 (talk) 14:23, 4 April 2014 (UTC)

I fear that there's no way to recover it. In future consider using Firefox Sync. Hunsu (talk) 12:05, 5 April 2014 (UTC)

How do people get a hold on e-mail addresses?

A couple of times, I received e-mails of job offers. They usually make the job too good to be true (high salary, low requirements, no resume/CV, no company identification). Often, I get suspicious of these e-mails, so I never reply to them. What irks me is the fact that these people somehow know my academic e-mail address, which I never publish anywhere or even use to receive subscriptions on any website other than school-affiliated websites, except ClusterFlunk.com, which requires an .edu web address, because it's a website created for students. Even so, I received spammy job offers even before I knew of ClusterFlunk.com. 140.254.227.76 (talk) 15:01, 4 April 2014 (UTC)

Do you have friends (assuming that you email them using the address you're talking about) who just can't stop clicking on things, opening strange attachments or installing Banzai Buddy-type stuff that random banner ads told them to install? That's often the way that the contents of people's email address books end up in the hands of spammers and scammers. --Kurt Shaped Box (talk) 15:22, 4 April 2014 (UTC)

i.e. BonziBuddy —Tamfang (talk) 05:09, 5 April 2014 (UTC)

Email address harvesting is a way of obtaining large numbers of emails for illegal purposes. Getting yours swiped is hard to avoid if you spend any time at all with email. The best way not to get burned is to remember, as you're doing, that if it's too good to be true, it isn't true. --Halcatalyst (talk) 15:27, 4 April 2014 (UTC)

University emails are usually pretty easy to guess. If you had a list of students from somewhere, you could churn through combinations of first.last@[university name].edu (or whatever) and most of them would be correct. APL (talk) 15:33, 4 April 2014 (UTC)

It's the same with corporate email addresses as well. Most companies follow a pattern. So just being familiar with that pattern is enough to get most people's email address correct. And to follow up on what was said above, many people use their work email address for personal business. (I see quite a bit of it doing tech support.) It wouldn't be surprising if one of those people also clicked on a link for malware which scraped their contact list. And finally, you don't mention if you are a student at this school or faculty. If you're faculty, your email address is likely listed somewhere on the school's web pages. Dismas|^(talk) 15:53, 4 April 2014 (UTC)

It's also worth googling your email address (within quote marks) occasionally to see if it's being harvested from a website somewhere.--Shantavira|^{feed me} 10:33, 5 April 2014 (UTC)

April 5

Lock out bad area on HD?

I ran HDTune's error check on an old external HD and it showed a damaged spot. I ran chkdsk /f and then I ran a full format. I ran HDTune again, and it did not show the damaged area. The bad area has been locked out, right? Bubba73 ^{You talkin' to me?} 01:53, 5 April 2014 (UTC)

One comment: When one area goes bad, others often follow soon, so you might want to back up your critical data elsewhere. StuRat (talk) 15:32, 5 April 2014 (UTC)

This is my oldest external drive. But it is used for backups from the main drive, so if it dies, nothing will be lost. But it might be about time to retire it. Bubba73 ^{You talkin' to me?} 16:18, 5 April 2014 (UTC)

how to set "plugin container to firefox" to start only upon my permission

overwhelmingly, the problems with my PC's behavior are due to "plugin container to firefox" related to "Adobe Flash Player" (as shown in task manager) -- almost every site has some god damn ad in a small screen, ready to slow down my computer for things I don't want -- THANKS TO FREAKIN "plugin container to firefox" related to "Adobe Flash Player" -- WHICH, EFF YOU

what do I do? I open task manager and end "plugin container to firefox"

I want to be able to have an alert letting me choose whether or not I allow "plugin container to firefox" to activate

how? how can I make "plugin container to firefox" be something that EACH TIME IT WANTS TO ACTIVATE I can choose whether or not to allow?

[and to those of you who work for browser firms -- OH MY GOD DO I HATE THIS]76.218.9.50 (talk) 08:44, 5 April 2014 (UTC)

Flashblock -- Finlay McWalterᚠTalk 09:23, 5 April 2014 (UTC)

NoScript is pretty cool, too. InedibleHulk (talk) 22:58, April 8, 2014 (UTC)

Broadband speed

I've recently changed ISP to BT (in the uk). I've just run two broadband speed tests. BT.com reckon download is 38.4 Mb/s, upload 1.82 Mb/s. Whilst http://www.broadbandspeedtest.co.uk/ reckon download is 8.663 Mb/s, upload 1.532 Mb/s. The uploads I can believe, but how can the downloads be so different? What are they measuring? -- SGBailey (talk) 09:16, 5 April 2014 (UTC)

It's possible that BT is telling you the ATM synch speed between your property and the DSLAM. This is the maximum theoretical speed that you would see if nobody else was using the internet. The speed test is showing you the speed you're achieving between your home and their server, and this will be restricted by other user's traffic at busy times. It's rather like the M25: the upper legal limit is 70 mph, but when there's a lot of other traffic, you can never achieve that.--Phil Holmes (talk) 09:55, 5 April 2014 (UTC)

Speed test sites work by downloading a large file to your computer and measuring the rate at which it is transferred. It would make sense that you could download things faster from your own ISP because you are directly connected to their network. Still, it's possible broadbandspeedtest.co.uk's connection is overloaded. I'm seeing a slow speed reported on that site, as well. So I would try downloading a large file from another site and observing the download speed in your browser's downloads window. For example, this file.—Best Dog Ever (talk) 10:06, 5 April 2014 (UTC)

You could try a broadband meter such as tbbMeter. It will record a graph of how your speed changes over time.--Aspro (talk) 14:44, 5 April 2014 (UTC)

Is there a way to do this?

I'm not looking for specific lines of code here -- I'm eager and excited to do that myself. But before I start writing code, I wanted to check if this would actually be possible.

What I want to have is a table for movies listing what films I have. There will be various columns for things such as "title" "director" "genre" etc. My question is,

Is it possible to make a sortable table (like, "click the column header to sort by that column") where one entry has more than one value in a given column.

Example: A movie like Alien I'd want to have two genre entries -- one for Science-Fiction and one for Horror. When sorting by title, I'd like it to appear once as: Alien Ridley Scott Dan O'Bannon Science-Fiction Horror

BUT

when sorting by genre, I'd want Alien to appear twice. Once under Horror, and once under Science-Fiction.

Thoughts?

My initial plan is to write all of this in XML and use XSLTs to do the viewing (I have no access to a database/server) Thanks! 199.94.70.36 (talk) 15:59, 5 April 2014 (UTC)

It certainly is possible, yes. Depending on the language, etc., it might be easier to put in two (or more) entries for each and filter out the duplicates where you don't want to see it repeated. Also, you might want to not list all the movies by all genres like that, as it will make for a rather long list. Instead you could select a theme, and list only movies that match that. StuRat (talk) 16:17, 5 April 2014 (UTC)

yeah I was planning to have each <movie> element have two <genre> child elements if I want to assign the movie to two genres, but I wasn't sure if I'd be able to still have that "clickable column" thingie under the conditions I set. This is actually intended to be viewed by OTHER people and I know many people want to be able to view by genre199.94.70.36 (talk) 16:25, 5 April 2014 (UTC)

We should also talk about the data structure. If you only have one field type with a variable number of entries, then you can put that last and not specify the number, just read until you hit the end of the line:

Alien, Sci-Fi, Horror

But, if you have more than one field type like that, (let's use main character's names) then we run into a problem:

Alien         , Ripley    , Sci-Fi, Horror
The Terminator, Terminator, Connor, Reese, Sci-Fi, Action 

Here we can't tell where the character names end and the genre's begin. We'd need to add the number of character names to clarify things.

Alien         ,1, Ripley    , Sci-Fi, Horror
The Terminator,3, Terminator, Connor, Reese, Sci-Fi, Action

Of course, we can add in the number of genres, too, if we want:

Alien         ,1,2, Ripley    , Sci-Fi, Horror
The Terminator,3,2, Terminator, Connor, Reese, Sci-Fi, Action

Or we could skip the numbers and use a delimiter instead (let's say a vertical bar):

Alien         , Ripley                   ,|, Sci-Fi, Horror
The Terminator, Terminator, Connor, Reese,|, Sci-Fi, Action

(XML files may have their own way to handle this problem.) StuRat (talk) 16:34, 5 April 2014 (UTC)

• Presumably the plan is to view this using a browser. Since browsers are generally not capable of handling arbitrary XML code, you'd have to use an XSLT to convert it to html. If you've figured out how to handle that, then you might find this page useful for an idea on how to make your table sortable. One way or another, it will require either writing or digging up some code in Javascript or some other language to do the sorting. Looie496 (talk) 19:15, 5 April 2014 (UTC)

You could do this really easily in Excel, if the XML stuff isn't all that important. Shadowjams (talk) 07:58, 6 April 2014 (UTC)

There is the aditional problem that a comma can appear within a movie title.

One could encode the offending characters somehow; I'd guess that for a comma within a field, you could invent some HTML-like tag, say, , to handle it in comma-separated lists.

Another idea, you could prefix every field with a field identifier, or a wiki-style namespace prefix. Say Ti for title (why? see below), Di for director, Ac for actor, Ch for character and Gn for genre.

Ti:Alien, Di:Ridley Scott, Ac:Sigourney Weaver, Ch:Ripley, Gn:Sci-Fi, Gn:Horror
Ti:The Terminator, Di:James Cameron, Ac:Arnold Schwarzenegger, Ac: Linda Hamilton, Ch:Terminator, Ch:Connor, Ch:Reese, Gn:Sci-Fi, Gn:Action

(Note that only the first colon of a field can be a namespace separator; it's fine and unambiguous to write, say,

Ti:Jaws 34: Revenge of the Alien Killer Sharks From Uranus: The Sequel of Jaws 33: Attack of the Alien Killer Sharks From Uranus

without risk of "colon cancer." Not sure if that's still relevant or if I'm going off a tangent again...) - ¡Ouch! (^{hurt me} / _{more pain}) 08:14, 9 April 2014 (UTC)

Wireless USB access, simplicity

I have a 2TB external USB drive I use for backup and storage of big files I need once in a while. What's the simplest standard for making access to it wireless, so my laptop backs up every hour plugged in or not? I have a wireless network, however I don't want others on the network to have access to its contents. I know Wireless USB does exist, but it didn't really take off and doesn't have the same recognition as Bluetooth, USB and so on. --89.243.7.60 (talk) 17:46, 5 April 2014 (UTC)

Strange flash problem

Resolved

 – 71.20.250.51 (talk) 14:50, 6 April 2014 (UTC)

All of a sudden, all flash games are running at hyper-speed.  This happened when playing a game when I was attempting to reset screen-size, and (presumably) unintentionally pressed some F-key that changed game speed, and I can't figure any way to restore this.  The strange thing is, this hyper-fast speed applies to all games from any source.  In addition, even videos on the flash player are fast.  I've already tried deleting cookies, and clearing the caches for Firefox, Adobe player and Java.  However, the problem persists.  Is there some way to reset key-bindings to defaults?  This is the only thing that I can think of to hopefully fix this.  Other suggestions are also welcome.  I'm using Firefox on Win7.  ~Thanks in advance; this is quite annoying and puzzling.  ~|71.20.250.51 (talk) 18:54, 5 April 2014 (UTC)

Sounds like it's using a higher frames per second rate. Can you try another browser to see if Flash games are affected there, too ? StuRat (talk) 02:29, 6 April 2014 (UTC)

Try restoring your computer to a date before the problem started.—Best Dog Ever (talk)

► I forgot the Windows' golden rule: "When in doubt, reboot."  The problem went away, but I wish I understood what exactly the problems was.  —71.20.250.51 (talk) 14:50, 6 April 2014 (UTC)

Searching Craigslist

When searching Craigslist, is there any option to have a "wild card" character in your search term? For example, in another program, if I use the search term "People Mag*" (with an asterisk at the end after the "g" as a "wild card") that will deliver in the search results any items that start with those letters (and end with anything else). So, that search would yield "People Magazine", "People Magazines", "People Mags", even "People Magic", and so forth. Is there a similar character (such as the asterisk) that works as a "wild card" for Craigslist searches? Thanks. Joseph A. Spadaro (talk) 20:40, 5 April 2014 (UTC)

The asterisk works for me. --  Gadget850 ^talk 20:58, 5 April 2014 (UTC)

Really? Does it work in this search ... Searchcraigslist.org ...? That is the search in which I am trying to employ a wild card. Thanks. Joseph A. Spadaro (talk) 21:07, 5 April 2014 (UTC)

You left that out. Odd. 'farscap*' gets hits for Farscape but 'farsca*' does not. Its a Google Search so it should conform to that standard. --  Gadget850 ^talk 19:13, 6 April 2014 (UTC)

Yes, it's odd. Any idea why that might be? I am trying to search for both People Magazine (singular) and People Magazines (plural "s" at the end). When I do a search for "People Magazine", it will not pick up the plural "People Magazines". And I also cannot get it to do the wild card asterisk, "People Mag * ". Any ideas? Thanks. Joseph A. Spadaro (talk) 22:29, 6 April 2014 (UTC)

April 7

Why would this store sell a Toshiba 10" 32GB tablet for only $2.70? And what dirt can you find about this store?

You see, I have a sinking feeling about this. Am I being lured into a trap? Or are they trying to liquidate their stock?

www.buydigitalproduct.com/toshiba-excite-at305t32-10-1-inch-32-gb-tablet-computer-wi-fi-nvidia-tegra-3-1-20-ghz?language=en&currency=USD $2.70 for THAT kind of tablet makes me wonder about its legitimacy.]

www.buydigitalproduct.com/index.php?route=information/contact This is where I find the store's supposed contact information. The next link proves that Google Maps doesn't know about "Jerry Dove Drive."]

Google Maps won't find Jerry Dove Drive anywhere in that town!

This WHOIS lookup doesn't provide this store site owner's info because some kind of infoguard keeps it hidden. Is this normal of any online store? Or would legit online stores allow their info to be made public on a WHOIS lookup?

I need help knowing more about this store before I enter sensitive details to buy this tablet for supposedly $2.70. Why do you think they'd price it THIS low? Hopefully you can help me figure out how genuine all this is. Thanks! --2602:30A:2EE6:8600:F9AF:6350:3C80:67FE (talk) 02:55, 7 April 2014 (UTC)

Do you have a link which lists that price ? I'm guessing they meant $270.00 for a used one, and messed up the decimal places. And I sure wouldn't send sensitive info to anyone you can't verify like this. StuRat (talk) 03:07, 7 April 2014 (UTC)

www.buydigitalproduct.com/toshiba-excite-at305t32-10-1-inch-32-gb-tablet-computer-wi-fi-nvidia-tegra-3-1-20-ghz?language=en&currency=USD This link] shows "Price: $476.99 $2.70 Ex Tax: $2.70." What does it show to you? I'll put up a screenshot in case you don't see what I see. --2602:30A:2EE6:8600:F9AF:6350:3C80:67FE (talk) 03:32, 7 April 2014 (UTC)

I get the same site as Nil. StuRat (talk) 14:53, 7 April 2014 (UTC)

Can't speak for StuRat but for me your link redirects to [1] which has no price and just directs you to Amazon to buy it for around $255. Actually this applies to pretty much any link to buydigitalproduct.com.

In any case I've never heard of prices that low excluding mistakes even in a liquidation sale. Perhaps $50-100. As always, if it sounds too good to be true and the site has several question marks (like the business behind it doesn't even seem to exist), there's no reason to think it's anything but what it seems i.e. a scam.

I don't know how much the shipping costs but if you want to throw away $15 or whatever the total ends up being I guess go ahead. But you may want to ensure the site is using a reputable payment processor and you only enter your credit card details there, otherwise you need to keep a close eye on your bill and hope your credit card provider doesn't make too many problems refunding you for fradulent transactions even though you entered your details in to a website which was an obvious scam.

Nil Einne (talk) 03:40, 7 April 2014 (UTC)

SCREENSHOT OF MY VIEW. Why does it redirect to someplace else for you? Anyway, I found it on the Froogle shopping search. See the listing that says "$2.70 from Buy Digital Products?" That led me to the page I've been trying to show you. I get a gut-feeling that a "came from Froogle" cookie instructs the site NOT to redirect to that Juni place, but to show me what I had intended to see. But what do you make of this? --2602:30A:2EE6:8600:F9AF:6350:3C80:67FE (talk) 04:26, 7 April 2014 (UTC)

By the way User:Nil Einne, I would've used a debit card. How different are fraud protections on a debit card if attempting to buy this tablet turns me into a fraud victim? --2602:30A:2EE6:8600:F9AF:6350:3C80:67FE (talk) 04:29, 7 April 2014 (UTC)

No, visiting the site through Froogle doesn't change anything. Presuming the site still shows the same page for you even when you refresh, the more likely possibility is it uses geolocation although this doesn't really explain StuRat's result presuming they had the same thing (but perhaps it's more specific than just the US). Alternatively or in addition, perhaps it's based on browser type or similar.

I obviously can't speak for your specific credit or debit card, but debit cards often do have lesser legal and perhaps policy based protections. See e.g. [2] [3].

Perhaps a key point when it comes to giving your number to a site almost certain to be a scam is that with a credit card is you can often refuse to pay for transactions which are in dispute. So while you may lose access to your credit card, at least you hopefully don't lose thousands of dollars temporarily. If some misuses your debit card, the money is gone from your account and you're often SOL until the matter is resolve even if you're sure to win. (If you only have a small amount in your account, you may get the same protection if your account is pushed in to overdraft or you may not.)

Either way even if you don't mind whatever loss for some small transaction for a site which is obviously a scam seeking such details, and even if you're sure you will be protected from fraudulent transactions you didn't authorise even though you voluntarily handed over you number to such a site, I don't think it's ever a good idea to do so, whether it's a credit or debit card.

(Of course as I mentioned earlier, even if the site is dodgy, provided you are sure you only hand over such details to a reputable payment processor your only concern should be the transaction itself. This is something a lot of people seem to miss.)

Nil Einne (talk) 05:07, 7 April 2014 (UTC)

I get the same negative result, and no, we do not need to see a screen shot. I've also removed all the live links to the commercial site. The URLs are still there so anyone who cares can still see the pages in question. WP:EL says we should not provide advertising links and even if this is an honest inquiry, that's what these links are. If the junustore is legit, we should not be linking to it and if its is not legit, even more so. As for the broader question, it could be an error, since corrected. Otherwise our article Internet fraud discusses many types of scams.--agr (talk) 04:10, 7 April 2014 (UTC)

Does a "Time Machine" for Wikipedia exist?

There's currently this Wikipedia art project in progress, where they're going to print out wikipedia at a snapshot and arrange it into an encyclopedia to physically visualise the scope of the wikipedia project, updates would be delivered real time by an old 'neverending print roll' type printer.

Which gave me an idea - is there a current way to browse wikipedia at a default date? so you could search for articles at a current point in time and for it to come up instead of going through the history tab? I think this would make for an interesting educational and fun tool in the future to show how understanding changes.

I don't know if you've ever thumbed through old encyclopedias, but it's a fun and enlightening experience - it'd be nice to do the same virtually.

Wikipedia has been around since January 2001, as an example, the world trade center didn't even warrant a mention until October. Little things like that.

Does this exist now? I had a look at the special pages and I couldn't see anything particularly relevant.

If it does not - where would be the best place to put in a 'feature request'? — Preceding unsigned comment added by Posty (talk • contribs) 03:47, 7 April 2014 (UTC)

There is the Internet Archive. It lets you browse a variety of sites at the time they were indexed. You could also just download a database dump of Wikipedia from this site and just set up your own outdated mirror.—Best Dog Ever (talk) 04:29, 7 April 2014 (UTC)

Wikipedia:WikiProject User scripts/Scripts/TimeTraveller doesn't work for me. Wikipedia:WikiProject User scripts/Scripts/TimeTraveller.js was last edited in 2009 so maybe it needs maintenance. PrimeHunter (talk) 04:37, 7 April 2014 (UTC)

I'm not sure it's actually correct that the World Trade Center "didn't even warrant a mention until October" of 2001. Notice that on that old-version page both the "next revision" and "previous revision" links actually point to the same version. And if you go to the article history and hit "oldest", you will see that the entry corresponding to that version shows the size reduced by tens of thousands of bytes from a previous version that isn't there. Compare, say, the article on the current baseball season which was created last year; if you look at its oldest version, the "previous revision" link is grayed out, and the revision history shows a positive increase in size with the first version. I think there must have been some technical change in October 2001 which is causing the early history of the WTC page to be missing. --50.100.193.30 (talk) 11:22, 7 April 2014 (UTC)

Yes the early page histories are sometimes problematic. There's brief mention of this problem at Help:Page history and Wikipedia:Usemod article histories but Wikipedia:Wikipedia's oldest articles probably has the most info.

As mentioned there, because of the software at the time, the history wasn't always reliably preserved nor was it always imported. The importation of old history has improved over time and a backup of some of the history feared lost has been found but I think some is still missing. And perhaps more importantly, a fair amount of the old history has AFAIK not been imported. (And for stuff imported or moved between earlier versions of the software, it didn't always work properly.)

In particular, for the stuff in the backups that didn't survive to the versions of the database used for imports, I'm not even sure if it's in directly browseable condition or you'd need to download the (fairly small) archive and view it yourself. And it's perhaps worth noting that while stuff until August 17, 2001 is supposed to be preserved in some fashion, it's not clear to me where this leaves the history from there until UseModWiki was replaced in ?'early 2002' ? (There are obviously copies of later versions of UseModWiki versions, e.g. the Nostalgia wikipedia [4] but perhaps these have been imported in to some other form or are simply preserved as front end HTML and the diff_log and rclog may be lost unless someone finds a backup? The email [5] is somewhat unclear to me as one the one hand it seems to be announcing the discovery of backups but on the other hand it later mentions how it was discovered that the history which was thought often deleted by the software, was actually preserved in one form so it was present in these backups.)

BTW, the Sep11 wiki was AFAIK started to deal with stuff considered unsuitable for wikipedia at the time [6] and the attacks were I think something of a catalyst as the first major current event dealt with by wikipedia. While I don't know what was accepted on wikipedia and what was pushed to the September 11 wiki, it seems unlikely the article on the WTC was created in October. (I'm presuming the OP is referring to the NYC WTC.)

Nil Einne (talk) 12:46, 7 April 2014 (UTC)

"Until a few years ago, such a project would have been impossible." Around 2005 I worked for a company that specialised in "short run, mono, multi-volume" print jobs. These were mainly regulatory drug applications, running to 400+ volumes, 12 copies. I considered at the time printing WP and donating it to the BL (of course WP was smaller then). A print run like this would have taken abut two weeks. All the best, Rich Farmbrough, 11:46, 8 April 2014 (UTC).

But, would you keep it up to date? - ¡Ouch! (^{hurt me} / _{more pain}) 08:40, 9 April 2014 (UTC)

Python

I have a text file which stores a single tweet per line. Each tweet is separated from the next tweet by a blank line. I wrote the following bit of code in Python to read the text file, store the tweets in a list variable, and print all the tweets which start with an "R" (retweets). However, everytime I run this, the code runs okay for a part of the text file and then stops abruptly and gives me a "list index out of range error". Can anyone point out to me where I'm going wrong?

tweet_db=[]
with open('C:\Users\La Alquimista\Desktop\hindistream2.txt','r') as f:
	for line in f:
		if line!=' \n':
			tweet_db.append(line)
			tok=nltk.word_tokenize(line)
			if tok[0]=="RT":
				print line

Thanks in advance. =) La Alquimista 14:55, 7 April 2014 (UTC)

If you could upload hindistream2.txt to a site like pastebin, so we could try it ourselves, that would be helpful. -- Finlay McWalterᚠTalk 15:21, 7 April 2014 (UTC)

Although surely the problem is that some line tokenises to an empty array, so tok[0] is out of range. -- Finlay McWalterᚠTalk 15:23, 7 April 2014 (UTC)

Could that error be trapped and handled properly ? StuRat (talk) 15:40, 7 April 2014 (UTC)

@Finlay: Thanks! All I needed to do was insert a line if tok!=[] to debug it. :) La Alquimista 17:03, 7 April 2014 (UTC)

Unrelatedly, you should double the backslashes in the file path, or insert r before the open quote (making it a raw string), or replace the backslashes with forward slashes. Your example happens to work in Python 2 because \U, \L, and so on are not valid escape sequences, but it will fail in Python 3 where \U is a Unicode escape, and it will fail in Python 2 (probably with an unhelpful error message) as soon as you try to open tamilstream.txt or whatever. -- BenRG (talk) 20:48, 8 April 2014 (UTC)

New Google maps

I use google maps which has recently ?updated automatically so that I have a different map presentation. There are a cluster of icons and buttons in the lower left corner and a black square in the lower right corner. They all offer an invitation to click by a change of cursor but they are all inactive. I cannot find any way to bird's eye view or street view. I have followed "The Tour" intended to explain the new format but it is almost incomprehensible offering no answer to my needs. I can zoom in and out and that's about it. What can I do to either go back to the old format (if that's possible) or to properly install the new version. With thanks in anticipation. Richard Avery (talk) 15:38, 7 April 2014 (UTC)

Do you have Java enabled ? If so, maybe try a reboot to make the icons pickable. StuRat (talk) 15:44, 7 April 2014 (UTC)

I haven't seen this new version, but the old Google Maps requires JavaScript enabled, not Java. --50.100.193.30 (talk) 08:45, 9 April 2014 (UTC)

No icon of the little man to click and drag? Towards the bottom right, near the zoom in and out control, is the icon that looks like a question mark talk bubble for help. Maybe it's behind the black square, but maybe for you it's in the cluster of icons on the left side that I don't have. The very last option in the help pop up menu is "Return to classic Google maps." However, it does sound like you don't have some sort of plug in running right. If you went back to the old style, it may not work either. A few work arounds you have try that can tell us what to try next is. Install another browser that you don't have. (Firefox, Chrome, Opera, IE, Safari) Try it again and it should load any plug ins it needs. If that works, your browsers plug ins should be reinstalled. Until the fix is found, you could also install Google Earth on your PC for free. Try it, you'll like it. Wonderley (talk) 05:50, 8 April 2014 (UTC)

OK, thanks both, that all sounds embarrassingly obvious, I'll work though that. Richard Avery (talk) 06:07, 8 April 2014 (UTC)

Upgrading my Linux computer to a new dual-boot computer

I am running Fedora 17 Linux on a computer that is already more than half a decade old. My workplace offered to give me a more powerful computer for home use, but the thing is, it runs Windows 8, and the company wants to keep it that way, so I can use it for work as well. Now the computer has two HD bays but is only using one of them, so I agreed with my boss that I can install the drive from my old computer to it as well.

Here's how this is supposed to go: I install my old computer's HD to the vacant bay in the new computer and connect both drives. I then upgrade the Linux installation from Fedora 17 to Fedora 20, installing a dual-boot system in the process. When the computer powers up, it gives me a choice of OS to load. If I leave it alone for a few seconds, it loads Linux, but if I choose so, it loads Windows instead. The whole Linux installation and the boot loader are on my old HD. The new computer's existing Windows HD must be left absolutely intact. I would also very much like to keep all my personal files on the old Linux HD, luckily they are all on a separate partition from the system (/) and boot (/boot) partitions.

Is this thing possible? If so, then how? JIP | Talk 18:12, 7 April 2014 (UTC)

A common way to get around those kinds of problems is to boot the computer from a USB memory stick. If you have the BIOS set up to attempt to boot from USB first - then from the Windows drive second - then merely unplugging your USB stick will return the machine to "normal" instantly. What goes onto the USB stick is something like GNU GRUB that you can set up either to unconditionally use your Linux drive - or to come up with a menu that offers you the choice of booting into Windows or into Linux. I'm not 100% sure what settings you need - but I'm fairly sure it's possible, Grub is a very flexible piece of software, designed to solve exactly this kind of problem. SteveBaker (talk) 19:45, 7 April 2014 (UTC)

I assume by "return to normal" you don't mean "it's suddenly displaying your Windows desktop", which was how I first read it. Presumably, you would shut the machine down, pull out the USB, and restart, in that order, correct?

Anyway, the problem with booting from USB is that it takes a loooong time. If you're one of those folks who likes to brag about their uptime, maybe you don't care -- start the boot, go out to lunch, and then leave it powered up for the next year. But if you do care, you'll probably want to configure your BIOS to boot from your old drive instead. It should be possible to have GRUB on the old drive, and allow booting into Windows as one of the options. You'll likely have to fiddle with it a bit, and I don't guarantee it'll work. --Trovatore (talk) 20:07, 7 April 2014 (UTC)

A Windows 8 machine will (almost certainly?) have Unified Extensible Firmware Interface rather than a BIOS, so in addition to changing the boot order, it'll probably be necessary to set it into "compatibility mode" (or whatever it's called) so it will an boot unsigned loader like GRUB. -- Finlay McWalterᚠTalk 21:18, 7 April 2014 (UTC)

Cripes, always something new you gotta keep track of ... when did that start? I don't remember having to do anything special about it when I put my latest laptop into a dual-boot config, about two years ago. --Trovatore (talk) 22:01, 7 April 2014 (UTC)

It's my understanding (second hand, having never done it myself) that machines which ship with Windows 8 must have UEFI, which is preconfigured with MicroSoft's cryptographic public key, the private counterpart of which is used to sign Windows 8's bootloader. The most recent version of Ubuntu, and presumably Fedora, also has a signed bootloader. Without that, one needs to turn off the check in UEFI, which makes it work like a legacy BIOS. I honestly don't know if an OEM Windows 8 will boot in that circumstance. An alternative to doing this, which involves commercial software that again I've no experience of, is discussed in this article. Were I in JIP's position, I would not dual boot - I would install VirtualBox on the Windows machine and install Linux in that. -- Finlay McWalterᚠTalk 22:20, 7 April 2014 (UTC)

It's probably time to buy a new drive anyway, put your old drive in a USB enclosure and use it for back-up. All the best, Rich Farmbrough, 11:50, 8 April 2014 (UTC).

That would be easier than trying to upgrade the Linux installation and replace the bootloader, I could just install an all-new copy of Fedora 20 on a fresh new HD and then copy all my personal files across. But how do I go about putting the old drive in a USB enclosure? I know my way around USB, IDE and SATA cables, but I couldn't solder anything or make my own cables or anything like that to save my life, even if I had the full instructions in front of me. Are there any ready-made USB enclosures that I could just plug my old drive's SATA cable into available? JIP | Talk 18:04, 8 April 2014 (UTC)

RegEx

tweet=re.sub('[http://\w*]','',tweet)
tweet=re.sub('http://\w*','',tweet)

What difference will the extra square brackets make in the code, if any? La Alquimista 21:12, 7 April 2014 (UTC)

Please take a look at the syntax section of our regular expression article. In the first example, the square brackets define a set of possible matching characters ('h', 't', 'p', ':', '/', any "word" character, or literally '*'), only one of which needs to match a single character in the string being tested to trigger a substitution. The second example is likely closer to what you want, matching a sequence of characters. Note that "\w" matches any "word" character - typically upper and lower alphabetic characters, digits, and the underscore. The combination "\w*" matches zero or more such characters. I suggest that you try one of the online regular expression testers like http://www.freeformatter.com/regex-tester.html to see how they behave. -- Tom N talk/contrib 23:55, 7 April 2014 (UTC)

So if I wanted the regular expression to match only a URL (nothing before that, nothing after that), then my code should look something like this?

tweet=re.sub('http://[^ ]*','',tweet)

What is the difference between writing the any-character-except-a-space part as [^ ] and as [^\s]? Much thanks. La Alquimista 10:21, 8 April 2014 (UTC)

I believe "\s" includes \f\n\r\t\v, but it may depend on the implementation and switches. Certainly \t, in everything I've used. All the best, Rich Farmbrough, 11:54, 8 April 2014 (UTC).

Those two regexes you gave are wildly different. I'm pretty sure you don't have a good grasp on regex La Alquimista. your first example is a single character, required... your second is a leading whitespace character, maybe.... your earlier question is even more confused. what exactly are you trying to do? Shadowjams (talk)

@Shadowjams: Heh heh. You're right. I'm very new to regex, and need it only for part of my code, so I'm unwilling to teach myself all the nitty gritties right away. What I want to do is take a line of text and delete any urls (starting with http:// or https://) from it. La Alquimista 09:24, 9 April 2014 (UTC)

April 8

Name that electric switch

I want to have some rather bright lights go on when I open the garage door. The garage door opener has it's own provision to do that, but is limited to 60 watts. I want 200 watts. So, when the line coming from the garage door opener is energized, I'd like that to flip a switch to turn on my lights using mains power, and then turn it back off when the garage door opener line goes dark. What's the name of the switching device to do this ? StuRat (talk) 03:51, 8 April 2014 (UTC)

Any commercial product that did what you wanted would be called a switch, but it would actually be some sort of Relay.

https://en.wikipedia.org/wiki/Relay

One that is either triggered by AC electricity and you plug it into power AND the light socket of the garage door opener with a socket to outlet converter as the switch.

https://www.google.com/#q=socket+to+outlet+adapter

Or, a relay that is switched by light. But you would have to make sure the additional lights you added would not trigger the light sensor to your switch. It could make for unusual positioning.

I've looked and I'm not seeing any kind of product. However, the 60 watt limit on garage door opener is due to the heat that would discolor or melt the white garage light cover. You could plug in a socket to outlet converter (mentioned earlier) and run an extension cord to more powerful light. However, I would use LED bulbs and maybe not go quite as bright as 200 watts. With LED bulbs you are getting the additional brightness while drawing considerably less electricity and not generating as much heat through the garage door wires just to play it safe.Wonderley (talk) 05:33, 8 April 2014 (UTC)

I'm using 15 CFL bulbs at 13 watts each, for a total of 195 watts, to light up the entire exterior of the house, for security reasons. StuRat (talk) 06:18, 8 April 2014 (UTC)

I did that too. I used a motion detector. As the relay. A car coming up the driveway signals it. I wired all that up 20+ years ago. I think I got the motion "switch" at Home Depot.Wonderley (talk) 16:09, 8 April 2014 (UTC)

I recommend wiring up a mechanical relay that is closed by the power for the opener light. Katie R (talk) 17:37, 8 April 2014 (UTC)

And now I see Wonderly mentioned that - skimming the response I thought it was just suggesting a light-triggered relay. You might also be able to find some useful signals on a terminal block on the opener. When I installed mine, I'm pretty sure there was a screw terminal with a TTL-level signal indicating the status of the light. That would give you a cleaner looking installation, and you could still have a bulb in the opener. Katie R (talk) 17:43, 8 April 2014 (UTC)

I have a fancy power board called a master slave, i wonder if you could use something like that? For example you plug your PC into the master and everything else (monitors, speakers, external peripherals etc...) into the slave sockets. The board monitors the current in the master socket and only turns on the slaves when the master is drawing then a few hundred mA. So if you plugged in your roller door into master and all your other lights into slave, that might work, depending on the layout of the socket etc... Vespine (talk) 06:16, 9 April 2014 (UTC)

How much does that cost ? StuRat (talk) 08:26, 9 April 2014 (UTC)

Great idea Vespine! That's the perfect relay. Hmm,...if it works. I have one of these for my PC.

http://www.walmart.com/ip/Inland-NuGiant-Energy-Saving-Smart-Surge-Protector/14664815

When power from my PC is off (The PC is plugged into the surge protector's outlet that works as a switch), it kills all power to the other outlets on the surge protector (Monitor, printer, speakers and could do more). These are the other surge protector's "Energy saver" outlets. Plug this product (Or similar) into the power outlet normally used for the garage door opener. Plug the garage door opener into the one AC outlet on the surge protector outlet for a PC or TV (The switch). Plug your lights into the "energy saver" (Switched) outlets.

A PC uses power even when it is off, but the surge protector knows it's just a little. I suspect the little bit of electricity used to operate just the remote receiver is little enough to be recognized as "off". But just the additional lights has to be recognized as "on". I'm sure the motor will be seen as "on" as far as the surge protector is concerned, but it runs for a very short time. If the lights in the garage door opener do not draw enough power to make the surge protector think it is "on", it will only make the lights plugged into this surge protector only work while the door is in the process of opening and the motor running.Wonderley (talk) 10:40, 9 April 2014 (UTC)

protecting a folder

How can I protect a folder in Windows 7? Thank you.175.157.16.134 (talk) 10:06, 8 April 2014 (UTC)

Protect it from what ? Being read by others ? Modified by others ? Copied by others ? Deleted by others ? Being read/modified/copied/deleted by viruses ? Accidental modification or deletion by you ? Hard disk failure ? StuRat (talk) 13:03, 8 April 2014 (UTC)

Your question is unclear, but if you right-click on the folder name (in Windows Explorer), then click "Properties" you will be offered various security options.--Shantavira|^{feed me} 14:39, 8 April 2014 (UTC)

You can protect your folder from external threats by downloading the free application . The application is open-source i.e., you can see what you are doing BUT with certain limitations. You can download it here . The size of the file is just 200 KB. or You can search Stegnography in wikipedia — Preceding unsigned comment added by Shrimadhav (talk • contribs) 08:59, 9 April 2014 (UTC)

More intense disk activity

Some time ago I found this useful utility - http://download.sysinternals.com/Files/ProcessMonitor.zip - and got my machines behaving quite nicely. Now there is a lot of disk activity, which happens mainly while the screens are in power-down mode. There also seems to be a lot less free disk space than I thought. Is this some new un-feature from a Microsoft upgrade? Or is it a known security issue I should worry about? I am fairly security concious, but of course visitors jump on and off the LAN these days, so my security model is a little outdated. All the best, Rich Farmbrough, 11:26, 8 April 2014 (UTC).

Depending on your operating system, it might be the indexer for Windows Search--Phil Holmes (talk) 14:27, 8 April 2014 (UTC)

Thanks, I did mean to say that indexing is turned off. Who knows though with Windows... (Vista)> All the best, Rich Farmbrough, 15:16, 8 April 2014 (UTC).

I have a 15" CRT monitor and intend to buy a 1GB GPU which would give 1900X1200 resolution but would my monitor support it?

I don't care how tiny things would look!!! I don't care how tiny things would look!!! I don't care how tiny things would look!!! I just want to play games in better graphics however i can lower down the screen resolution while playing the game. . . . currently i have 1024X968 resolution on a NVIDIA 6150SEn430 GPU and i want a better GPU and i've played Bioshock, Halo 2 and few others games on it but in low graphics mode :(( I have processor: AMD Athlon Dual core 64X2 5400+ (additional info) and 3Gigs of RAM.(additional info) . . LG500G is the model of the monitor — Preceding unsigned comment added by Shikka Kun (talk • contribs) 13:38, 8 April 2014 (UTC)

Every monitor has a maximum resolution that it will support, and there is no way to go beyond that. For yours the maximum resolution is 1024x768. If you want better resolution, you'll have to buy a better monitor. Looie496 (talk) 14:03, 8 April 2014 (UTC)

And that likely means a larger monitor. Alternatively, you could use bump scrolling, if that graphics card supports it, to view a portion of the scene at once, giving you a larger effective resolution, although you only see 1024x768 of it at a time. If you are on Windows 7, 64 bit, you can use Microsoft's own magnification feature to do this. At a 200% zoom level, for example, you'd get a total resolution of 2048x1538, but would only see 1/4th of the screen area at once. StuRat (talk) 14:08, 8 April 2014 (UTC)

BTW, if the expense is the reason not to get a new monitor, you can get a used CRT monitor for free or almost free, as everyone wants a flat screen now. StuRat (talk) 14:24, 8 April 2014 (UTC)

And there are surprising bargains on older LCD's, even 1080i LCD TV's which support the resolution you need. We heard what you said (3 times) but really, the electricity you would spend on running a CRT (assuming you get a cheap LCD) would soon exceed what you are trying to save. Sandman1142 (talk) 22:14, 8 April 2014 (UTC)

Freeview Manual Tuning - Missing Channels

I was looking at the wiki page for the Freeview channels for the UK & I noticed that there where a few channels listed on that page which aren't listed on my TV's channel list. I auto retuned & when I've looked again at the TV's channel list The missing channels aren't there. Does anyone have a list for the UHF Channels for the TV channels on Freeview so I can do a Manual Retune ? 194.74.238.6 (talk) 14:11, 8 April 2014 (UTC)

If it's like in the US, we had lots of digital TV channels that came and went in short order, so the channels may really be gone now. StuRat (talk) 14:19, 8 April 2014 (UTC)

Manual tuning is unlikely to help. A whole multiplex is transmitted in one radio channel, so you'd expect to get all the channels in the multiplex or none. And if the autotune didn't see a strong enough signal, the channel will be too weak to watch (digital signals don't degrade as gracefully as their analog forebears). What's in which multiplex is listed at Digital terrestrial television in the United Kingdom#Multiplexing. The frequency a multiplex is broadcast varies by transmission tower - so you need to figure out which tower your antenna is pointed at. See {{Television transmitters in the UK}} and sites like http://www.thebigtower.com But if you're not getting all the multiplexes your tower transmits (check: some towers or relays don't carry all yet) you might need a better antenna. In a few cases people with older Freeview equipment needed to change the firmware in their digibox or tv to handle changes; that should have happened automatically, but you might want to see if you can force that. -- Finlay McWalterᚠTalk 14:27, 8 April 2014 (UTC)

Oh, right. Well for example, on the Freeview (channel list) page CBS Reality & on the page for CBS Reality, it on Freeview channel 66. If I had/got the UHF Channel for CBS Reality I could try & retune my TV to see if I can find it. 194.74.238.6 (talk) 14:32, 8 April 2014 (UTC)

I've checked & I think I'm under the Winter Hill transmitter. 194.74.238.6 (talk) 14:34, 8 April 2014 (UTC)

I know that, for example, BBC channels (I think all of them) are UHF Channel 50 (at least for me). So like I say if I could get a list for for all the TV & radio channels & what UHF channel they're on I could try & retune manually.194.74.238.6 (talk) 14:38, 8 April 2014 (UTC)

The channels, and frequencies, for Winter Hill are listed at Winter Hill transmitting station#Digital. -- Finlay McWalterᚠTalk 14:55, 8 April 2014 (UTC)

Thanks, I'll go & see if they can help. 194.74.238.6 (talk) 15:07, 8 April 2014 (UTC)

I just retuned my TV (I'm on a different transmitter, so different UHF frequencies to you) and I do indeed now get CBS Reality at #66 on the EPG. As the articles you linked indicate, it is on the COM4/SDN multiplex (same as ITV3+1 and ITV4+1 and other garbage channels). Winter Hill transmits that multiplex on UHF58. I would expect that if you have antenna problems you wouldn't see anything in that multiplex. -- Finlay McWalterᚠTalk 15:17, 8 April 2014 (UTC)

openning tex file in swp 5.5

Scientific work place 5.5 is installed on windows xp in my computer .My supervisor has sent me a thesis files through mail, which contain a tex file abc.tex and other files as title.tex ,abstract ,preabble1.tex and bibliography.tex ,when i try to compile abc.tex message appear ,"valid .cst file not found ,document may not load correctly.would you want to continue ." when i continue this then it opens,then when i try to compile and create pdf file, no pdf file is created,in ttue tex DVI previewer it says ,preamble1.tex not found .I have searched it on internet ,i have found that problem is with tex style file.Then I imported all files in new folder then try,then while compiling no pdf is created ,because it can not find title.tex,abstract.tex ,..... . Style of this thesis is similar to harvard. — Preceding unsigned comment added by True path finder (talk • contribs) 14:30, 8 April 2014 (UTC)

Somewhat ironically given your user name, it seems that the problem is that the LaTeX processor does not have its search path set up correctly. The main file abc.tex probably consists largely of lines saying \include{foo} for various values of foo, but when the processor looks for the file foo.tex, it doesn't find it. Unfortunately I don't know enough about your setup to go beyond that. Looie496 (talk) 15:03, 8 April 2014 (UTC)

Text messages on Nokia Lumia

I recently needed to send a text message to a recipient on my mobile phone. The phone I have is a Nokia Lumia 820 my company gave me to use. I didn't have the recipient's number already in the contacts list on the phone, so I tried to just send the message to the actual number, like it had been possible on all previous Nokia mobile phones for the past two decades. Disappointingly, I didn't find any way to do this. The text messages application only allows selecting recipients from the contacts list, and the only actual numerical telephone keyboard I could find on the entire phone (in the telephone application) only allowed me to actually call the number, or store it as a contact. Is there any way to actually send a text message to an actual number on Windows Phones? It has been possible on previous, non-Windows, Nokia phones for over two decades, and I am sure most of these new-fangled touch-screen phones from other vendors (not running Windows) allow it too, although I'm not sure, since I've never used them. Has no one in either Nokia or Microsoft ever thought of this? JIP | Talk 18:46, 8 April 2014 (UTC)

According to the user manual (link) within the 'create new message' area you simply type the number in (suggestion being it's the same field you would add a recipient to). Relevant page of the manual I've linked to is page 47. Sorry it's not a phone/system come across but I am 99% certain it will be easy, it's just modern phones have made a good job of making easy tasks more difficult by their desire to be minimalist in on-screen design. ny156uk (talk) 19:40, 8 April 2014 (UTC)

If that works, it's the same for Android phones at least with the default non Hangouts SMS app. In Android, if you're desperate for a numeric telephone keyboard for some reason, you can I guess temporarily switch your keyboard to one although it seems easy enough to just use the QWERTY keyboard if it's only one or two recepients. I've never had that much of a problem entering a phone number using the QWERTY keyboard personally, even on my old LG P500 with its small screen although I guess it may be a bit easier if you have a large number of recepients you want to enter by phone number. Alternatively with Android you could I guess download an alternative SMS app which would provide an option for a numeric keyboard when entering the recepient. Perhaps there are other options, I'm not sure since as I mentioned I've never felt the keyboard to be a big problem. And I can't say which, if any, of these will port over to Windows. Nil Einne (talk) 06:41, 9 April 2014 (UTC)

April 9

Controlling the amount of time a piece of code executes itself

I'm fairly new to Python, which is why I've been going around asking all of these questions. I've cooked up a bit of code that streams tweets from Twitter in real time. So once I run the following code, the tweets keep streaming in and saving themselves to a text file on my desktop. The thing is, I haven't figured out a way to end the script yet, and I resort to closing the entire Python window to stop streaming. I'd like to put this into a function that I can call, but that would be cumbersome if there is no way to keep the script from running infinitely. I could either specify the number of tweets after which to stop, or the time elapsed after which the function stops. Please tell me how I can set a small timer that stop the execution of the function after, say, 5 minutes? How do you even create time variables in Python? And what is the format for specifying time?

# -*- coding: utf-8 -*-
import json
import time
import tweepy
from tweepy import Stream
from tweepy import OAuthHandler
from tweepy.streaming import StreamListener
ckey = 'UgSJvQnCduRRCJegx58Gw'
csecret = 'piMz329KUX5SPP2KQePKQo6w0qo9EAgkja78KAMwBo'
atoken = '45792734-rYCswBixWZilYMJMrVZ1wlz0dGR4Kltvh46844xpd'
asecret = 'aJ7gw9J5NpSglHgd3PuuUYqdAfiMRGifoAYA6V83eZ61P'
class  listener(StreamListener):
    def on_data(self, data):
        try:
            data = json.loads(data)
            tweet = data['text']
            print tweet
            with open("C://Users/La Alquimista/Desktop/hindistream.txt", "a") as f:
                f.write(tweet.encode("UTF-8"))
                f.write('\n\n')
            return True
        except BaseException, e:
            print 'failed ondata,',str(e)
            time.sleep(5)
    def on_error(self,status):
        print status

auth = OAuthHandler(ckey, csecret)
auth.set_access_token(atoken, asecret)
twitterStream = Stream(auth, listener())
twitterStream.filter(track=["को".decode('utf-8', 'ignore')])

La Alquimista 09:30, 9 April 2014 (UTC)