Wikipedia:Reference desk/Archives/Computing/2013 June 26

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Computing desk
< June 25 << May | June | Jul >> June 27 >
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.

June 26[edit]

Copying large files from PS3 HDD?[edit]

I've downloaded a couple large videos from the Playstation Store (the PS4 announcement video, 6GB, and the E3 Press Conference, 5GB). Like all game videos from the Playstation Store, they are just plain h264 movies without DRM. But Playstation 3 only seems to support FAT32 for USB attached external drives. And I get an error when trying to copy these files because I assume they are larger than the 4GB file size limit on FAT32 volumes. Are there other ways to get files off the PS3? Perhaps through the ethernet port? --Navstar (talk) 01:00, 26 June 2013 (UTC)

People have been asking this question for years on Sony forums and other various sites. The only method suggested that hasn't been confirmed yet (but seems worth trying IHMO) is using the PS3's built in browser to upload the file onto a file hosting site like Mediafire or Rapidshare (I reccomend Mediafire) and then you can download the file onto your PC from the site. I think that could work; you could try that. --Yellow1996 (talk) 16:03, 26 June 2013 (UTC)
Unless you have fast upstream broadband, it would be faster to run an http file-upload server on the target machine, such as this one. (Edit to add: or this one, which has no GUI but doesn't even need to be installed.) -- BenRG 01:02, 27 June 2013 (UTC)
I pointed my PS3 browser to Dropbox. When I go to upload a file, the PS3 only lets me browse the Photo directory on its internal drive. There doesn't seem to be a way to browse other directories on the local drive. And unless I can choose that file, it doesn't matter if I use Dropbox or a HTTP upload server on my home computer. --Navstar (talk) 03:12, 27 June 2013 (UTC)

I installed Ubuntu on my HP laptop.[edit]

After installing it on my Laptop I really did not like it as much as the Windows it came with out of the factory, I wish to re-install Windows how would I do this because I can barely read the activation code on the bottom, how do I reinstall windows again? — Preceding unsigned comment added by (talk) 04:51, 26 June 2013 (UTC)

If you have the original media (or a set of recovery disks) you will not need the activation code. I've just done a reinstall on a hp 6730b and a 6710b and neither required a code. (talk) 06:45, 26 June 2013 (UTC)
Ubuntu is like windows, in so much, that the first time you encounter it there is a steep learning curve before you get the hang of it. Linux is Not Windows However, preserver for no more than the 'same time' that it took you to learn windows and you will find Ubuntu (or Mint or any other the other popular Linux distribution) way easier to use and maintain than windows. Then, in a year or two's time you'll be posting here to ask “why don’t more people use Linux? It's a breeze.” It may only be that you have come across windows first, and so a better operating system now seems strange. Had it been the other way about, you would not even give windows (with all the hoops it requires you to jump though) a first let, alone second look. --Aspro (talk) 14:36, 26 June 2013 (UTC)

I dont have the disc so how could I do it? — Preceding unsigned comment added by (talk) 07:00, 27 June 2013 (UTC)

Did your laptop come with a 'legitimate' installation of windows on it or did you buy a Grey Market laptop.--Aspro (talk) 12:11, 27 June 2013 (UTC)
It may have a recovery partition, although you may have wiped that out when installing Ubuntu. HP Support also used to sell replacement discs for $12, if I remember right. (talk) 12:23, 27 June 2013 (UTC)

On demand web graphics[edit]

Can people suggest software that is good at creating on-demand graphics for web servers. In my particular application, we have large data files and we are looking for ways to extract and plot small subsets of that data in response to user requests submitted via a web interface. What existing software packages are good at such applications? The web interface itself will probably be PHP, but we can of course call other programs to generate the images as needed. Dragons flight (talk) 05:40, 26 June 2013 (UTC)

For general drawing, people use ImageMagick (libmagick), cairo, and GD. To do mathematical and statistical diagrams and plots there's things like matplotlib, GNU Octave, and gnuplot. There are also lots of charting libraries (which produce histograms, pie charts, and the like) such as pChart. -- Finlay McWalterTalk 10:38, 26 June 2013 (UTC)
There's another option you may consider: doing the plotting client side (in the web browser). There are several JavaScript charting libraries (which render to a Canvas element, or in some cases do things like pie charts and histograms with CSS). That way, rather than generate a PNG or SVG server side and send that, you can send the data (e.g. as JSON or XML) and render it in the browser. That can lower the amount of data you have to move on the network, but the big gain is you can make the render responsive and interactive - so the user can select what data to view, zoom around, turn labels on and off, etc. The downside of that is that it depends on the user having a newer browser. -- Finlay McWalterTalk 12:44, 26 June 2013 (UTC)
  • I think this depends to a great degree on the format of your data files and the types of information you want extracted. If they are spreadsheets or can easily be turned into spreadsheets, you might find that the easiest solution is to write macros for a spreadsheet program. Looie496 (talk) 14:49, 26 June 2013 (UTC)

Recommender system in wikipedia[edit]

Hi ! I am currently working on recommender systems and someone mentioned to me that wikipedia had one and I can't find any mention of such thing though. Am I just blind and wikipedia does recommend articles to users depending on what they have already read ?

From what I understand, the "See also" section is filled out by users right ? Or is there any kind of system that recommends the articles which need to be in that "See also" section ? Thanks for your help ! Fperrotti (talk) 06:58, 26 June 2013 (UTC)

There's just the links in the articles. Wikipedia doesn't set a curriculum, so it's up to individual readers to set their own course through the articles. Sometimes the category links will lead you other places, and sometimes the portals on the page will have subject suggestions. Wikipedia content is sometimes linked from external things (like Wikibooks) which do set a curriculum, but the authors of a specific Wikipedia article can't know why someone came to read it, and so can't recommend where you'd go from there. The "see also" section certainly isn't a recommendation list, although it is sometimes misused as one - it should be links that are relevant to the article but that can't easily be worked into the article's text (where they'd have context) - an example of good practice is Extreme points of North America, which see-alsos other extreme-points articles. -- Finlay McWalterTalk 10:31, 26 June 2013 (UTC)
Wikipedia has an article about Recommender system about the subject, it doesn't operate one. Wikipedia also has articles on subjects like singular value decomposition which I'm surprised that article doesn't seem to link to but you do sometimes have to click around a bit on the links in an article to see what meat if any there is in Wikipedia about a topic. Dmcq (talk) 11:43, 26 June 2013 (UTC)
SuggestBot suggests pages that may be of interest to an editor based on their edit history - so it's a bit like a recommender system. But I don't think anything in Wikipedia logs pages that you have looked at (as opposed to actually editing them) so it can't make recommendations based on what you have read. Gandalf61 (talk) 12:07, 26 June 2013 (UTC)
An important point about Wikipedia is how it is structurally different from sites like Amazon and Facebook. Those deliver a personalised web experience; most visitors have an account and most of those are signed into it most of the time. That account means Amazon and Facebook know who their visitors are and deliver custom content based on their preferences, actions, and history. In contrast Wikipedia serves almost everyone the same content, with only a statistically negligible number (signed in editors) getting any kind of custom content. Amazon's and Facebook's value all lies in that individualised experience, but it comes at a massive cost to them. Wikimedia makes very effective use of cache (web cache and memcached), meaning it can serve the whole site with 1000 servers or so (I can't find an up-to-date number; it was 300 in 2008), but Amazon and Facebook both have a number of gigantic datacentres - Facebook spends half a billion US dollars a year on servers (ref). To properly support visitor-specific content over the whole visitor base at Wikipedia would mean abandoning that massively cache efficient architecture and adopting an Amazon-like architecture. So, handwaving the numbers a bit, implementing a recommender system at Wikipedia's massive scale might cost upwards of $200 million US. Donate here!! Smiley.svg -- Finlay McWalterTalk 14:22, 26 June 2013 (UTC)
Wikipedia does provide an implementation of an expert-system that recommends other articles: Special:Book has a "suggest a page" feature. This is implemented by the MediaWiki Collection extension and is not very well documented. A little bit of tutorial information is provided at Wikipedia Help for Books. Nimur (talk) 21:27, 26 June 2013 (UTC)
Maybe the OP is thinking of the suggestions Wikipedia makes as you start to fill the search box in. For example, if I type in "m", it suggests Mollusca and Mexico and if I add an "a" afterwards, it switches to Marriage and Major League Baseball. Somehow, those articles are being recommended to me, though I'm not sure how (number of hits, perhaps?). Matt Deres (talk) 16:43, 27 June 2013 (UTC)


Hi. When I created my Gmail account, I entered my real name in the registration form. The problem is that Gmail uses this real name as sender in the e-mails that I send to other people. How can I set Gmail to use an alternative name as sender instead? I can't find this option in Settings. Please don't tell me to create another account because I like my current account name. Thank you. -- (talk) 15:23, 26 June 2013 (UTC)

Follow this tutorial (with pictures!) --Yellow1996 (talk) 15:53, 26 June 2013 (UTC)
It works! Thank you very much. You are wonderful. -- (talk) 16:51, 26 June 2013 (UTC)
No problem! I'm glad it worked out for you! :) --Yellow1996 (talk) 18:44, 26 June 2013 (UTC)

Where can I find the Matlab Psychophysics toolbox?[edit]

I'm looking for the Psychtoolbox-3, but is down. Does anyone know where it might be elsewhere located? Thanks in advance! --Rajah (talk) 16:02, 26 June 2013 (UTC)

You can find the source code at, but you would need some sophistication in order to use it. You can find an older version with a Matlab download script at, but I don't guarantee that it will work for you. Looie496 (talk) 16:23, 26 June 2013 (UTC)
Just as a side note, works fine for me. There's a really useful site I know called downforeveryoneorjustme which allows you to check if a site is down for everyone, or just you. --Yellow1996 (talk) 18:49, 26 June 2013 (UTC)
It works now. Earlier it was showing an invalid https security certificate. Looie496 (talk) 21:51, 26 June 2013 (UTC)
Thanks everyone! It's back up now as you pointed out. I appreciate all your help! --Rajah (talk) 05:15, 30 June 2013 (UTC)

Excel: mixing time and currency[edit]

Hi. I'm trying to use Excel to do timesheet work. It's a mix of cells relating to time and money.

I have managed to get it Excel calculate, based on a start and finish time how many hours and minutes have elapsed, so for example, 09:45 and 10:15 in columns A and B renders 00:30 in column C.

So far, so good, but when I try to get Excel to combine column C with column D (rate per hour) I get nothing but gibberish in column E, which should be the chargeable amount, based on time and rate.

Can someone give me a formula for column E that will work? Thanks in advance. --Dweller (talk) 21:39, 26 June 2013 (UTC)

try =C1*24*D1 but make sure the cell D is formatted as general format - excel might try to convert it to a datetime format which isn't correct. --nonsense ferret 22:37, 26 June 2013 (UTC)
Terrific, thanks that worked, even when I formatted D as currency. --Dweller (talk) 08:31, 27 June 2013 (UTC)

Help setting up website[edit]

I recently set up a new internet forum, which proved rather easy, since all I had to do was download some free software and install it on my website, but now I want to expand the site, to include pages for people to share pictures and maybe even video relating to the topic of the site, would this be as easy to set up, or would I need some sort of professional help? Anyone want to point me in the right direction, is it something I could arrange myself with no experience of actually writing websites, it's a pretty common feature, I think, just a case of getting my own copy, and then somehow linking it to my existing site... (talk) 21:57, 26 June 2013 (UTC)

Depending on which forum-making package you used, there's probably an extension/addon available for it that can do what you need. Check and see if they have an official site or community support forums. --Yellow1996 (talk) 22:03, 26 June 2013 (UTC)
worth thinking carefully about what you are trying to do. Do you mean you want to allow people to upload pictures/videos to your forum website to share them from there? If so that would obviously have quite a large impact on the bandwidth that your site uses, and the amount of hosting space that you would need. For this reason some forums take the option of allowing users to embed files hosted by other webservices such as youtube. Doing so might reduce a few headaches of hosting files on your own site such as copyright/ bandwidth and security. --nonsense ferret 22:44, 26 June 2013 (UTC)
  • We know from an earlier question that the OP is using Go Daddy. I'm pretty certain that Godaddy won't allow you to use your web site for third-party file sharing. Generally it costs quite a bit to get a service that will, and even then there are going to be limitations. Looie496 (talk) 22:47, 26 June 2013 (UTC)
I've run two forums on GoDaddy. Both with image galleries that I hosted on the site. The images could be hot linked from the galleries onto other sites.
OP, there is likely some other software package that you can download and install rather easily but without knowing more about what software you are already working with, it's hard to tell you more. Also, you may want to look into a CMS or Content Management System if you intend on having non-forum pages as well. Dismas|(talk) 23:05, 26 June 2013 (UTC)
I'm running Vanilla Forums at the moment, I can look around their website some more, but I haven't seen any add-ons for this so far. And yes, I know I'll need a lot more bandwidth and storage space once the gallery starts growing, I'm willing to pay what I need to for that. (talk) 14:00, 27 June 2013 (UTC)
There's a Vanilla Forums addon called FileUpload which looks like it's exactly what you need. --Yellow1996 (talk) 16:34, 28 June 2013 (UTC)