Wikipedia:Reference desk/Archives/Computing/2010 February 13

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 13

Ubuntu format taking forever

I've plugged an external 1.5TB hard disk into this PC here and booted from the latest Ubuntu install disk. I went through the install wizard and the "Installing system" dialog is now showing, claiming "Partitions formatting". It's been stuck at 5% for a really long time. Is there a way to figure out whether it's dead or not? Comet Tuttle (talk) 00:28, 13 February 2010 (UTC)

run iotop (or better yet iotop -a); you should see a mkfs.ext3 process chucking out a fair amount of write iops; no writes means no progress. -- Finlay McWalter • Talk 00:52, 13 February 2010 (UTC)

Thanks for the tip — iotop isn't installed, so it recommended I do a sudo apt-get iotop, but then apt-get didn't recognize iotop as a package; so off the power went! Thanks - Comet Tuttle (talk) 01:05, 13 February 2010 (UTC)

As can be seen here iotop is in the universe section of ubuntu repositories. Universe isn't enabled by default so you would have needed to enable it to install iotop. --194.197.235.240 (talk) 13:12, 13 February 2010 (UTC)

Google Buzz

I've been sucked into Google Buzz just like a horrific number of other people, and, because I don't like the idea (I will save my reasons for another site as Wikipedia is not a soapbox) I want to get out from it. Is there any way that I can exit this intrusive project that came along to my email account with no warning nor any information whatsoever on how to use it? TIA! --KageTora - (影虎) (A word...?) 01:36, 13 February 2010 (UTC)

First off, go to Settings/Labels, and you can hide the "Buzz" label. If you want to completely filter out Buzz, see this Lifehacker Article: Hide/Remove Google Buzz Updates from Your Gmail Inbox. -^Avicenna_sis @ 03:10, 13 February 2010 (UTC)

It's not the hiding/removing of the updates that I am concerned about. I don't want people following me. [soapbox] As a case in point, when I opened gmail a couple of days ago and found that I had this GoogleBuzz added to my account, I was automatically already following 6 people, completely without my consent. One of these people was a person who had contacted me a while back about translation of Japanese manga - we talked for a while, but the project fizzled out for various reasons and we both eventually forgot we'd even met. Anyway, now, GoogleBuzz is inviting me to go and look at pictures of his family that he put on Picasaweb (and emailed me the pictures anyway even if I didn't want to have a look), and he probably doesn't even know that GoogleBuzz has done this. This is why I don't want to be part of it and completely out of it. It's a privacy thing. Is there any way to just disable the ability for people to follow me? Also, I've written on here about Facebook and the complete lack of control that you have over people writing whatever they want on your 'wall', and I am thinking that this will become the same. I already know the names and personal interests of many of the friends of the six people whom, by default, I am supposedly following, and really, I don't want my boss looking around and seeing what my mates, and their mates, and then THEIR mates are into. Ridiculous! Email and who I talk to should be my own business and no-one else's. [/soapbox] So, is it possible to disable it completely? --KageTora - (影虎) (A word...?) 07:15, 13 February 2010 (UTC)

I noticed this too, and I can't find a way to disable it either. I suggest writing to google and threaten to leave them for another email provider if they don't rectify the situation. If enough people don't want it and complain, they'll have to do something or risk losing their ad-revenue. This is also worth reading. —Preceding unsigned comment added by 82.43.89.27 (talk) 14:51, 13 February 2010 (UTC)

Cheers. That link shows there are plenty of people who have the same idea, though there is no information on how to opt out of this feature. Writing to Google may be the only option, and that is what I have just done, but, not surprisingly, in the drop-down menus where you have to select what service you are referring to, GoogleBuzz is nowhere to be found. 'Other' was the most appropriate one. We'll see how long this takes. Cheers. --KageTora - (影虎) (A word...?) 17:07, 13 February 2010 (UTC)

FWIW, I never clicked on any link connected with Buzz and only clicked the "Turn off Buzz" link. Now I can't find any mention of Buzz except for the link to turn it back on. Perhaps that;s the secret to maintaining your privacy. Astronaut (talk) 18:19, 13 February 2010 (UTC)

The first rule of Google Buzz is not to talk about Google Buzz. --Mr.98 (talk) 21:40, 13 February 2010 (UTC)

I believe (though I can't be sure) that if you haven't "connected" anything (like Chat status messages, or Google Reader or Twitter) to Google Buzz, no on will be able to see anything from you (if you have connected those things, you can remove them). (There's also a public profile might have set up.) So I believe that everything is private-by-default (except that other people can see that you exist; but the people who can follow you know that already). If my analysis is correct, Buzz is not inherently intrusive, it's just that Google should have been more clear. Paul Stansifer 00:10, 14 February 2010 (UTC)

Google has now added a way to disable Buzz completely. Also, it looks like I was wrong -- it used to connect to public Picasa and Google Reader items automatically. Paul (not signed in) 71.174.127.229 (talk) 17:11, 14 February 2010 (UTC)

Nice. Thanks for the update —Preceding unsigned comment added by 82.43.89.27 (talk) 17:46, 14 February 2010 (UTC)

Make server think I'm a browser

Resolved

Hello! I'm using Java to query a server (specifically, http://www.rae.es, a Spanish dictionary website), but if I do the exact same query (http://buscon.rae.es/draeI/SrvltGUIBusUsual?LEMA=tomar&origen=RAE&TIPO_BUS=3) in Firefox, then copy-and-paste it into my Java code, I get two different responses. For Firefox, I get the proper response to the query, but for Java, I get a response from the home page (http://www.rae.es). It seems like the server knows I'm using a automated program and is refusing my request. Is there any way to make the server think the request is coming from a browser, or otherwise fix this problem? Of course, I know about etiquette and will program a delay between queries. Thank you for any advice!--el Aprel (^facta-_facienda) 01:39, 13 February 2010 (UTC)

Try changing the user agent string. --Spoon! (talk) 02:59, 13 February 2010 (UTC)

Thanks for the tip, but googling "java change user agent" has only brought up dead threads on forums and incomplete answers. Can anyone help me with an implementation specific to Java? I mean, it should be as easy as calling a method like setUserAgent(String userAgent), but I haven't found something like this by browsing through the java.net package.--el Aprel (^facta-_facienda) 21:24, 13 February 2010 (UTC)

Assuming you're using URLConnection then an example is here -- Finlay McWalter • Talk 22:10, 13 February 2010 (UTC)

Thank you, Finlay McWalter. This is exactly what I needed.--el Aprel (^facta-_facienda) 03:50, 14 February 2010 (UTC)

Better performance tracking than htop

I'm using Kubuntu. Is a free utility available that can track CPU and memory usage for each program over the past few hours (rather than just the past second as htop does) and that can also track time spent reading to and writing from memory and disk, and ideally also GPU usage? NeonMerlin 08:02, 13 February 2010 (UTC)

Nagios is great, but may be overkill for your needs (it's usually for monitoring entire clusters, but can be used on one machine). Tracking GPU usage is a tough nut for technical reasons ("GPU usage" is a poorly-defined metric - do you mean bus traffic, GPU utilization, etc)... and it depends heavily on your hardware and drivers, so it's hard to get any utilities that track GPU usage well. Nvidia has some performance monitoring tips and a few tools in their developer area on their website - e.g., PerfHUD (unfortunately Windows-only). Nimur (talk) 20:38, 14 February 2010 (UTC)

Seeing the URL of an unclicked hyperlink in Firefox?

Something I found useful in Internet Explorer was that you could see the URL of a hyperlink, that you put the cursor over, at the base of the browser. Is there any way to get the same thing in Firefox? Sometimes the URL is shown in Firefox in a small yellow pop-up, but this is unreliable and you have to wait a long time for it, and it may just be a description rather than an URL. 92.29.55.65 (talk) 12:12, 13 February 2010 (UTC)

View --> Status bar. Friggums (talk) 13:36, 13 February 2010 (UTC)

TalkTalk ISP and eMail timeout.

I have a friend in the UK who uses the "TalkTalk" ISP - she is extremely computer-illiterate and a very slow typist. She says that: "after 20 minutes,my computer sent message to say "your > time has expired" while i was sending you an email and i had to type it all again". I'm guessing that this 20 minute limit is due to the nature of the "free broadband" service that they offer. I assume that there must be a way to compose email offline and use the 20 minute time limit to actually send it - but she's going to need step-by-step instructions on how to do that. Do any of you guys know enough about this service to give me a click-by-click account of what to do? I have no idea what email client she's using - but I strongly suspect that it's one that came with the TalkTalk service. Since I have no clue how TalkTalk works, any help would be greatly appreciated. SteveBaker (talk) 14:34, 13 February 2010 (UTC)

Well the obvious answer is simply to open Notepad, Wordpad or another text editor, compose the email in that then copy and paste into the email client when complete. Exxolon (talk) 14:43, 13 February 2010 (UTC)

If it is a broadband service, it will not expire. I'm wondering if she's using a webmail application, which has a session timeout? What she actually got was the mail service saying it had timed out, since she had not accessed any new pages? I'd suggest checking what she actually does for email - it may be possible to save drafts. But, as Exxolon says, she could compose using notepad and paste it in. --Phil Holmes (talk) 15:19, 13 February 2010 (UTC)

I have to correct Phil Holmes — I don't know anything about TalkTalk, but some broadband connections do "expire" — I used to be a Qwest DSL customer, and they intentionally disconnected my connection every 2 hours because that was the main inconvenience that incentivized customers to pay more for an upgraded plan. The ostensible reason was to "free up connections" for other users, but, suspiciously, they always assigned me the same IP address.... Comet Tuttle (talk) 16:08, 13 February 2010 (UTC)

I don't believe any broadband connections expire in the UK, which is where Steve was referring to. --Phil Holmes (talk) 11:34, 14 February 2010 (UTC)

Some modem/routers will also disconnect after a certain amount of inactivity (this can usually be turned off). In Malaysia time limited broadband (ADSL) packages exist Nil Einne (talk) 18:37, 13 February 2010 (UTC)

I'm a TalkTalk customer and my connection never expires/disconnects/timeouts due to inactivity. I would imagine it's as Phil Holmes says and her site session is timing out. TalkTalk do not package a standalone email client with their installation discs, they do offer webmail though at talktalk.com/mail. Nanonic (talk) 20:40, 13 February 2010 (UTC)

OK - so it's likely to be the talktalk.com/mail server. I'll try explaining how to cut/paste from notepad. Thanks guys. SteveBaker (talk) 23:24, 13 February 2010 (UTC)

No, not the mail server ("Message Transfer Agent"), but more likely the Webmail client, which has the role of a Mail User Agent. But copy&paste will solve it. --Stephan Schulz (talk) 20:07, 14 February 2010 (UTC)

Software Question

Hello, I was looking for a specific software that allows you to take pieces of different pictures and make it look like one. I don't want a morph program, I was thinking more along the lines of what they have at a plastic surgeons office where they could take your picture and change the nose or eyes to show how you could look if you got acertain procedure. Does anyone have any idea of what a program like that is called? —Preceding unsigned comment added by 71.137.255.212 (talk) 15:16, 13 February 2010 (UTC)

I don't know what you're referring to, but it does sound like morphing of some sort. --Mr.98 (talk) 20:39, 13 February 2010 (UTC)

there is a specialty software called FotoShop or PhotoShop that professionals use. It's used to make magazine cover girls have impossibly thin arms and legs, among other things. However, because it is specialty software it may be very expensive, maybe as much as several thousand dollars, I haven't looked. —Preceding unsigned comment added by 82.113.106.195 (talk) 20:27, 13 February 2010 (UTC)

Do you mean Photoshop? It's expensive, to be sure, but there are free alternatives (like GIMP, which does the same stuff). It isn't the kind of thing where you can just say "make them thin," it's just image editing software. It's not what the original poster is asking about, I am sure. --Mr.98 (talk) 20:37, 13 February 2010 (UTC)

The operation you are talking about is called photo stitching. It is usually used to make wide panoramic scenes. See photo stitching and panorama. Also, a simple Google search of "photo stitching freeware" returns many good free programs that can do this. —Preceding unsigned comment added by Roberto75780 (talk • contribs) 21:53, 13 February 2010 (UTC)

Any decent graphics software should be able to do what you want eg Paintshop Pro and FREE versions are available. This software won't do the job for you, it's more of a sophisticated electronic paintbrush. It requires some practice, and experience with other graphics software helps. Tools to "change the nose or eyes" are definitely included, these allow you to select an area and move it or alter its shape. Whether the result looks realistic is up to the users skill--220.101.28.25 (talk) 03:48, 14 February 2010 (UTC)

Article in TFA, ITN, OTD and DYK?

First of all, I did not post this in the help desk because, well, it does not deal with a technical problem on the Wiki. Now for my question: Has there ever been an article which was on the Main Page for TFA,DYK In The News and On this Day?

Regards,

Buggie111 (talk) 15:31, 13 February 2010 (UTC)

I'm not sure about OTD, since we don't seem to keep records of that on the talk pages, but for the other three we can take an intersection of Category:Featured articles that have appeared on the main page, Category:Wikipedia Did you know articles that are featured articles, and Special:WhatLinksHere/Template:ITN talk. Give me a few minutes. —Akrabbim^talk 16:17, 13 February 2010 (UTC)

Update: There are zero DYK-FAs that were ITN, let alone featured as TFA. Sorry to disappoint you. —Akrabbim^talk 16:26, 13 February 2010 (UTC)

That's impossible under the current rules, which disqualify articles that have been ITN for DYK. I am sure ITN has a similar provision. decltype (talk) 16:29, 13 February 2010 (UTC)

I didn't meen all on one day.

Buggie111 (talk) 19:50, 13 February 2010 (UTC)

We didn't either. DYK and ITN are mutually exclusive. Apart from that, there have been no FAs that were ever ITN, let alone TFAs. There have been 503 FAs that were DYKs, 216 of which were TFAs:

Sorry. I got it. Thanks.

Buggie111 (talk) 23:07, 13 February 2010 (UTC)

—Akrabbim^talk 20:17, 13 February 2010 (UTC)

Windows Media Player: removing the delay between repeated playings of a file

Might anyone be able to tell me whether I may remove the delay between playings of a file being repeated in Windows Media Player, such that it is continuously playing; and if so, how?

Thank you in advance,

88.189.248.66 (talk) 16:40, 13 February 2010 (UTC)

I don't think there is a delay on my computer. Perhaps your file needs trimming? Dbfirs 21:36, 13 February 2010 (UTC)

This is the file that I wish to have continuously playing; in between repeated playings, I have a delay which is as noticeable as it is short, whence my query. Do you hear it also? 88.189.248.66 (talk) 23:03, 13 February 2010 (UTC)

I tried with a trimmed file on my computer and noticed placing the same file on the playlist several times in a row results in a slightly shorter delay compared to the delay at the end of the playlist starting the list all over. —Preceding unsigned comment added by Roberto75780 (talk • contribs) 21:51, 13 February 2010 (UTC)

Thank you for your helpful suggestion! 88.189.248.66 (talk) 23:03, 13 February 2010 (UTC)

Yes, on checking carefully, there is always a short delay (sorry if I implied there wasn't - I hadn't tried white noise files). One solution would be to use sound editing software (GoldWave is shareware) to create a single file with many copies of the original sound end-to-end without a gap. The delay in Media Player would then not be a problem. Dbfirs 08:36, 14 February 2010 (UTC)

Connecting a DVD player into a computer

Can DVD players, in general, be used as a CD-ROM or DVD reader for a PC?--Quest09 (talk) 18:13, 13 February 2010 (UTC)

No. --Mr.98 (talk) 18:25, 13 February 2010 (UTC)

None of them?--Quest09 (talk) 18:29, 13 February 2010 (UTC)

(ec) In most DVD players the digital datastream is converted into composite video and analogue audio by the electronics in the player. I guess that someone with sufficient skills might be able to dismantle a DVD player and intercept the datastream, but it might not work and could permanently break a perfectly good DVD player. It would be a lot of trouble to go to when a computer DVD drive can be bought for as little as $15. Astronaut (talk) 18:30, 13 February 2010 (UTC)

No regular DVD players (for playing movies on TVs) are likely to have the requisite hardware to interface with a computer. They are fairly specialized devices. There is no reason they would have this capability built into them. --Mr.98 (talk) 19:45, 13 February 2010 (UTC)

awk problem

i have this script, where the input files have been sorted over $1 (using sort -k 1,1n in > out.sort), the problem is that is runs too fast, it makes only one output file of ~200k when i would be expecting 60 files each in the mb-gb region. I'm guessing its a problem with the next/exit statements as there the newest bits i added but i cant see whats going wrong. The idea is to split the input files into files of overlapping ra ($1), by looping through the file untill its close to the region is question, printing that out untill its out of the region then exiting and going to the next number in the list.

#!/bin/bash

for j in $(seq 1 1 60)
do
	ra=$j
	ARCSECMIN=$(echo "scale=6;$ra-0.002778" | bc -l)
	ARCSECPLUS=$(echo "scale=6;$ra+1.002778" | bc -l)
	
	PATHOUT="/path/out/$ra.ra"
	
	for i in "/path/in/*.sort"
	do
		awk '{ if($1 < ra1){next}
			else if ($1 < ra2) {print >> "'"$PATHOUT"'"}
			else{exit}
			}' ra1="$ARCSECMIN" ra2="$ARCSECPLUS" $i
	done

done

--86.27.192.94 (talk) 21:41, 13 February 2010 (UTC)

First, general shell script debugging: use the -x option. Add it to the #! line, or put set -x before the part that's not working the way you want.

Now the reason your script isn't working: you put the *.sort glob in quotes, so it's not getting expanded there. The inner for loop is only being run once, with i set to that literal string, including the asterisk. Later, when you reference $i, it gets expanded and globbed so all of the filenames get passed to single awk instance, and it treats them as if they were one big concatenated file. 98.226.122.10 (talk) 04:47, 14 February 2010 (UTC)