Wikipedia:Reference desk/Archives/Computing/2009 May 10

= May 10 =

Monitor settings and eye strain
I recently read somewhere an article discussing how certain settings on an LCD display monitor can cause eye strain. the article mentioned settings you can change that will reduce eye strain. I'm trying to find that article--or learn some other way what those settings are Rafael Mark (talk) 00:52, 10 May 2009 (UTC)

I found the answer (ot "an answer") elsewhere. The refresh rate of 60 hz bothers some eyes, including mine. I increased the refresh rate to 75 hz on my new monitor and the result was immediate. No discomfort to my eyes. I got this from a comment on another site, which is not authoritative enough to reference, so this should be considered merely anecdotal at this point. Rafael Mark (talk) 02:10, 11 May 2009 (UTC)


 * That is complete bogus, or in your case placebo effect. LCD displays, unlike CRT displays, are persistent. With CRTs the recommended refresh rate is 75Hz or above to scan the image more often on the screen so it flickers less, and causing less eye strain. LCDs have no such need and even if you can set the refresh rate to 1Hz the content of the screen will stay there, albeit only updating once every second. To reduce eye-strain try turning on ClearType (if you're on Windows, other OSes should have text AA enabled by default), and run your LCD screen at native resolution and set your dpi to something comfortable. --antilivedT 09:29, 12 May 2009 (UTC)

Scribd obfuscation
A search in Google took me to a page of Scribd, a site of which I'd never previously heard. When I clicked on the link I was told that I needed Javascript, so I switched to a browser in which Javascript was turned on. I was then told that I needed Flash, which I purposely avoid. This all seemed bizarre in that Google had presumably digested the strings I'd searched for without any help from either JS or Flash. I therefore clicked on Google's cached version and could read the text (in an old version of Opera, though only part of it in a new version of K-Meleon). (Judging from the frequent mid-sentence occurrences of the string S N L Varieties of Meaning plus what looks like a page number, I suppose it's a scan, legal or otherwise, of something titled Varieties of Meaning.)

As well as serving up nightmarishly humongous CSS stylesheets, it's clear that this site is doing a considerable amount of browser-sniffing and assorted dicking around. When I save the page, I can't find the text anywhere. This has tickled my idle curiosity. If anyone else here is interested too, I wonder what's going on here. -- Hoary (talk) 04:36, 10 May 2009 (UTC)


 * Changing your User-Agent to something like "Googlebot/2.1" or "Wget/1.11.4" will make it return plain text. There's also a Javascript-based menu at the top from which you can download PDF or plain text, though it seems to require logging in. This book may be a bootleg but it's definitely all-digital (perfect spelling, perfect alignment, no noise). However the plain-text extractor doesn't understand high-level document structure and just dumps everything on each page in rough top-to-bottom order. The S N L letters are in the margin of the odd-numbered pages for some reason. -- BenRG (talk) 14:02, 10 May 2009 (UTC)


 * Ah, good idea! In the meantime, I thought I'd try feeding the URL to the WDG's HTML validator, with "display input" turned on. Very, very bizarre results. No surprise that the web misdesigners optimistically had the DOCTYPE declaration cite XHTML 1.0 Strict, against which it abjectly fails to validate. (The maximum number of errors was reached. Further errors in the document have not been reported.) This degree of incompetence is normal for "professional" websites. But what gibberish it contains! Notably, row after row of a table, telling us e.g. error and mistake definition / barbara logic / scientific superstitious reasoning objectivity integrity concise intro logic / definition of language. Tomorrow I'll try Wget. Thanks again! -- Hoary (talk) 15:47, 10 May 2009 (UTC)
 * Scribd allows anyone to upload any one of numerous file formats for "on-line publishing" and has had complaints against it for pirating (including Harry Potter) because they don't check the ownership of the uploaded material. Some people might be uploading formats they don't quite handle, too, accounting for some of the gibberish. Alexa reports the average user spends 2.7 minutes there, so I don't see where they make their money. It's not on several suspicious-sites places that I checked.
 * @Hoary -- the images on it are all in Flash, so the text won't show up in Page Source.KoolerStill (talk) 19:37, 10 May 2009 (UTC)
 * "they don't check the ownership of the uploaded material" - yes they do; I've had a document incorrectly flagged as a copyright violation when it wasn't, so there's definitely a system running. Its effectiveness is another matter, but it's no more a haven for piracy than YouTube is. I like the service in general though. — Matt Eason (Talk &#149; Contribs) 20:02, 10 May 2009 (UTC)

I hav mini project in 8086 programming....i want to know that how to convert this C program to assembly language for 8086 ???
—Preceding unsigned comment added by 116.74.103.74 (talk) 05:22, 10 May 2009 (UTC)


 * One approach would be to use the GNU Compiler Collection along with the  option to compile the program to assembly language.  Note however that the generated assembly language would likely contain library routine calls to accomplish the I/O operations and possibly to manage the environment.  The generated code will also likely be much different than hand crafted code.  Also, gcc appears capable of generating code for the 386 family and above, but I did not see an option for 8086 mode, which is significantly different.  Perhaps an older version will have 8086 support.  -- Tcncv (talk) 05:54, 10 May 2009 (UTC)


 * Indented and spaced out the code to make it more readable. Astronaut (talk) 06:10, 10 May 2009 (UTC)


 * I recognize this as old Borland/Turbo C++ code. You could download Turbo C++ from here, and compile your program from the command line using the -S switch. The result is 311 lines of code, which of course make calls to the standard library. The result is a lot more readable than the ~1100 lines of code you get doing the same thing with another ancient compiler (Microsoft Visual C++ 6.0), but I really don't see the point of the assignment. I have in the past occasionally compiled single functions to assembly and tried to optimize the output by hand, but with a modern compiler, the compiler will probably do a better job at this task than you are able to. --NorwegianBluetalk 13:04, 10 May 2009 (UTC)


 * See http://webster.cs.ucr.edu/AoA/DOS/AoADosIndex.html. It has reference for some BIOS calls if you feel like using them and not external libraries. --194.197.235.70 (talk) 14:06, 10 May 2009 (UTC)
 * If the project is intended to teach you to write 8086 machine code - then using the compiler to do the job isn't going to help at all! Machine code written by the compiler looks utterly different than stuff you build by hand. SteveBaker (talk) 01:10, 11 May 2009 (UTC)

Google Gallery Script
I'm trying to use this script but it doesn't seem to work for me. script source

Background info: "Google Gallery Scipt" Description of the script (Archived link) Instalation link --Drogonov (talk) 10:59, 10 May 2009 (UTC)

Blocking Gmail Contacts
Is it possible to block some of my Gmail contacts from sending me emails. As in, suppose they try to send me something, the mail will automatically bounce back, and they'll receive a message saying that their mail wasn't delivered. Is there an option that can turn this feature on? 117.194.227.220 (talk) 08:55, 10 May 2009 (UTC) —Preceding unsigned comment added by 117.194.229.208 (talk)


 * No. You could set a filter that automatically deletes certain messages though. --grawity 13:21, 10 May 2009 (UTC)

Small Resolution Images
I was wondering how to scale a large resolution image, 3264 x 2448 pixels, down to 900 x 675 pixels while still maintaining a high quality image. This picture shows that is can be done: (click full-view). The picture I'm trying to scale is moderately high quality, little or no noise and has a lot of detail, but the detail and quality goes down a lot when I try to scale it using the cubic interpolation on GIMP. I assume that picture I used as an example was over 20 megapixel before scaling, so shouldn't it be easier to maintain quality on my smaller full resolution picture when scaling? BeefJeaunt (talk) 14:19, 10 May 2009 (UTC)


 * Cubic certainly looks OK to me, but have you tried the sinc interpolation option? — Matt Eason (Talk &#149; Contribs) 14:43, 10 May 2009 (UTC)


 * I can't find that option, all I have are none, linear, and cubic. BeefJeaunt (talk) 15:15, 10 May 2009 (UTC)


 * What version of GIMP do you have? I'm on 2.6.6 for Windows - I don't know what version sinc was added in — Matt Eason (Talk &#149; Contribs) 17:40, 10 May 2009 (UTC)


 * The secret is to sharpen after you downsample. Don't worry about cubic versus sinc versus Lanczos. -- BenRG (talk) 18:37, 10 May 2009 (UTC)


 * You could also try seam carving, which is available as a GIMP plugin. chocolateboy (talk) 21:09, 10 May 2009 (UTC)

Amount downloaded since reboot (Windows XP)
I downloaded a bandwidth-monitoring program (NetStat Live from AnalogX.com) which unfortunately didn't seem to like my network adapter, reporting zero bandwidth usage all the time.

However, it did accurately report the amount I had downloaded (and uploaded) since I last rebooted the computer, even though most of that traffic occurred before I installed the program. This suggests to me that Windows keeps a record somewhere.

I'm currently having issues contacting the author of the program, but figured I may as well ask here too -- does anyone know an obvious way to ask Windows how much has been downloaded and uploaded since the last boot?

Cheers, all. Rawling 4851 15:09, 10 May 2009 (UTC)


 * I know it's not exactly answering your question, but another good program for monitoring network usage is NetMeter, which stores all the bandwidth statistics in its program files directory. I've found it to be very reliable. 8I.24.07.715 talk 15:29, 10 May 2009 (UTC)


 * Thanks, I'd come accross that but not tried it 'til now; it's got quite a nice report feature. Interestingly, it only knows about the data downloaded/uploaded since I started it, not since boot. Rawling 4851 15:49, 10 May 2009 (UTC)


 * 1) Open Task Manager and click on the Network tab.
 * 2) Go to the View menu, and choose Select Columns. Turn on columns for Bytes Sent and Bytes Received. Click OK.
 * 3) Go to the Options menu, and check Show Cumulative Data. --Bavi H (talk) 02:31, 11 May 2009 (UTC)


 * Awesome, cheers, I only got as far as step 2 on my own. So Windows does indeed keep a record. Rawling 4851 13:01, 11 May 2009 (UTC)


 * Woot, I found a couple of C++ programs (FreeMeter and pcread) using performance counters to get the data I need, and managed to understand them enough to cut out what I didn't need. :D Rawling 4851 15:57, 12 May 2009 (UTC)

google and work internet
i work at a conservative workplace and sometimes not-so-conservative emails get sent to my gmail. I like to check my personal email at work, and this is certainly allowed, and i only click on emails i know will be ok. But...I'm afraid that gmail pre-emptively opens email that I haven't even read with its fancy scripting, and that technically my company's IS algorithms could interpret this as me reading "bad" things at work. any idea if this is just paranoia or am I onto something and if so is there a gmail setting I could turn off to help this? thanks, -Bob 173.30.14.113 (talk) 15:46, 10 May 2009 (UTC)
 * You can set gmail to always use https instead of http. You'll find it at the bottom of the settings page. I think this should eliminate the possibility of anyone intercepting the communication between your browser and gmail's server. The only minor disadvantage, is that if you use the iGoogle interface, the gmail gadget will no longer work. --NorwegianBluetalk 17:24, 10 May 2009 (UTC)
 * use Tor —Preceding unsigned comment added by 82.44.54.169 (talk) 19:24, 10 May 2009 (UTC)
 * HTTPS should solve the problem. Does Google transmits data preemptively to your web browser client?  Surely it does parsing and processing on all your emails at the server-side, but I was not aware it would start transmitting data to your webbrowser before you request it.  Unless it is doing such preemptive caching (maybe as some part of an "accelerator" technology), those unsavory emails will not ever be detected by your local network snoopers.  You should also be open about this stuff with your authoritarian IT people - they will probably understand that "random data" shows up on their network for a variety of reasons (have you ever looked at the output of a packet sniffer?!  WHOAW the things you see - 95% incomprehensible, 4% relevant, and 1% ... well, you know).  As long as it's not interfering with the workplace, hopefully nobody will care.  Nimur (talk) 19:47, 10 May 2009 (UTC)
 * "Unless it is doing such preemptive caching" - Yeah, it prefetches messages listed in the current view (certainly in the inbox - not sure about elsewhere) to speed up loading if the user opens the message. — Matt Eason (Talk &#149; Contribs) 20:30, 10 May 2009 (UTC)

cabling
advantages and disadvantages of twisted pair cables213.55.87.137 (talk) 16:39, 10 May 2009 (UTC)
 * We have this exact section in our twisted pair article — Matt Eason (Talk &#149; Contribs) 17:43, 10 May 2009 (UTC)

EEE PC 901
Does the EEE PC 901 come with out of the box support for 2gb of RAM? The Linux kernel I mean. I have searched the EEE PC wiki and it doesn't say that it definetly does. CAn anyone clarify?

Thanks 89.241.252.228 (talk) 18:28, 10 May 2009 (UTC)

Nevermind sorted it myself. Keep up good work 89.241.252.228 (talk) 18:57, 10 May 2009 (UTC)

OpenOffice clean startup
I have searched through the OpenOffice docs and forums, but I can't find this particular "feature" discussed anywhere. When I start OpenOffice (regardless of if I start writer, calc, etc...), it feels the need to open every OpenOffice document I've every created. It is becoming a huge nuisance. Right now, if I open any OpenOffice document, I have to wait a few minutes and then close a bunch of windows that I have no interest in seeing. This appears to be something related to sessions, but I can't find anything on OpenOffice sessions. Can anyone tell what this is called so I can forcefully disable it? -- k a i n a w &trade; 20:09, 10 May 2009 (UTC)


 * this describes how to do it, so maybe check and see if one of this has occurred and can be removed? 76.117.247.55 (talk) 00:35, 11 May 2009 (UTC)

disc synchronisation
What Mac OSX synchronisation software can synchronise entire drives, as opposed to merely folders? What I want to do is to keep the normal drive inside my mac in sync with an external hard drive - that makes a spare bootable drive. The difference between drives and folders is that synchronising drives is a simple matter of only one sync. To do the same for folders means making a separate synchronisation for each drive in the root folder, and a new sync, for each new folder/file added. 78.33.187.178 (talk) 23:50, 10 May 2009 (UTC)


 * List of disk cloning software has a few Mac OS X programs that will do what I think you're wanting to do. I don't know of any software that will actually "synchronize" two drives in the sense of preserving the most recent changes made on both drives.  Tempshill (talk) 02:31, 11 May 2009 (UTC)

Disk cloning software would work, but I would say disc cloning is rather different to synchronisation. When I used Carbon Copy Cloner to make an incremental backup, it took about an hour. It felt rather less "fault tolerant" than File synchronization - quite a list of software.78.33.187.178 (talk) 10:02, 11 May 2009 (UTC)


 * Some of the disk cloning software on that list - well, Norton Ghost is the only one I am familiar with - will make incremental backups of entire disks. Can you explain what it means to "feel" less fault-tolerant?  Tempshill (talk) 22:36, 11 May 2009 (UTC)

So can Carbon Copy Cloner. Maybe I'm being a bit paranoid about it, but when it does do an incremental (one hour isn't bad, actually) you sort of get the impression if you need to go, and you stop it half way through, you'll blow the backup. And on that note, maybe I ought to try it to see, I do happen to have the disc space for it.78.33.187.178 (talk) 23:49, 13 May 2009 (UTC)