Wikipedia:Reference desk/Archives/Computing/2007 October 16

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

October 16

SHA-512

I'm looking for a program that can create SHA-512 hashes on OS X (locally, no web apps). Are there any that aren't a pain in the ass to install? Cheers. --MZMcBride 01:01, 16 October 2007 (UTC)

Well... one half-way-there solution is to use PHP installed locally (which is very easy to to do on OS X—it often is already included) and run it off of the localhost as a "local" web app, and use its mhash functions, which seem to support SHA-512? I don't know if that would work for your needs (since I don't really know them), but it would be very easy to set up. --24.147.86.187 02:47, 16 October 2007 (UTC)

I am not sure but you probably already have the shasum script. Just open up a terminal and type shasum --help and see if it is there. -- Diletante 04:12, 16 October 2007 (UTC)

No, shasum --help doesn't work. Frankly, I'm shocked that no one has created an OS X app for this yet (or at least I haven't been able to find one). --MZMcBride 05:10, 16 October 2007 (UTC)

It looks like you can compile openssl to do SHA512 digests, according to the OpenSSL documentation. —Preceding unsigned comment added by Jsbillings (talk • contribs) 13:19, 16 October 2007 (UTC)

Routers

I have some sort of a D-Link Router set up in my home, sharing an internet connection with a wireless adapter thing on one computer and a LAN cable going to another. It worked perfectly when I first started using it but then one day the computer hooked up wirelessly just stopped being able to connect to the Internet. I was also using the Wi-fi on my PSP and it stopped working as well. It said there was a DNS error. However, the wireless network status states that the connection to the router is fine, and the one connected with the LAN cable is still working. It's really frustrating. Does anyone know how this would be fixable? Thanks Mix Lord 04:42, 16 October 2007 (UTC)

Try the following: unplug and replug the power cable on the router, then go to start, click Run, type in CMD. When the command prompt pops up, type /ipconfig release, then when that is done, type /ipconfig renew. This should reset essentially everything and fixes most problems related to networking. -Wooty [Woot?] [Spam! Spam! Wonderful spam!] 05:47, 16 October 2007 (UTC)

Okay I'll try that but if the connection with the router is still fine (which it seems to be) will that change anything?Mix Lord 00:29, 17 October 2007 (UTC)

information technology

hello sir/madam, my self lovenish mittal,me persuing my P.G. in M.Sc(comps.) from panjab university chandigarh. i have to give seminar on "PAST,PRESENT ANT FUTURE OF INFORMATION TECHNOLOGY" and after searching a lot me not able find out the material as per my teacher's requirement. she wants me to cover each and every thing changed til now in IT sector,its growth,inventions like networking,internet,hardware-software devlopement etc..as IT is very vast topic but still m confused from where to get all this.if u can help me out i will b very thakful. and please mail me your answers on '*********@gmail.com'. i wil waiting for your feedback. —Preceding unsigned comment added by 220.227.11.194 (talk) 05:13, 16 October 2007 (UTC)

E-mail address removed. Do not provide contact information, such as your e-mail address, home address, or telephone number. Be aware that the content on Wikipedia is extensively copied to many websites; making your e-mail address public here may make it very public throughout the Internet. Lanfear's Bane 09:10, 16 October 2007 (UTC)

You could start with History of computing, but please remember we are not here to do your homework for you. 87.112.85.54 12:03, 16 October 2007 (UTC)

Remember that 'information technology' is technology used in direct association with information - it's storage and dissemination- in that sense it includes paper, printing press, telephone/vision, library, etc... that should get you started.87.102.12.235 18:13, 16 October 2007 (UTC)

Faster Browsing

I was given advice to do this in Firefox; in the url box type about;config, then go to the filter text box type network, scroll until you get to network.http.pipelining; change that to true, and change the one below that from 4 to 10. Now always one to want faster, I tried it then tested my speed using [1], I did not get that different results. But I would like to know what this actually does, and what other editors think about this. Phgao 14:14, 16 October 2007 (UTC)

I think pipelining causes Firefox to download more than one element of the web page at the same time in parallel. This will not increase your maximum speed as measured by speedtest.net, but it may make web-pages load slightly faster in some situations. (If you (or the server) have a slow connection like a dial-up, this setting will probably slow page loading times.) 69.95.50.15 15:09, 16 October 2007 (UTC)

MozillaZine explains. Turns on HTTP pipelining. --h2g2bob (talk) 15:37, 16 October 2007 (UTC)

Thanks guys! Phgao 16:57, 16 October 2007 (UTC)

To the ip; why don't you consider getting an account :) ? Phgao 16:58, 16 October 2007 (UTC)

Latest OSS Software

what is the latest open source software? —Preceding unsigned comment added by Opie kriss (talk • contribs) 15:07, 16 October 2007 (UTC)

(add title) --h2g2bob (talk) 15:38, 16 October 2007 (UTC)

I'm not sure what you mean. The latest open source announced project? Who knows—dozens must be started every week. The latest finished application? (Can any of them be really described as "finished"?) The latest version of any given application? (So many to sort through, and what would be the point?) I think it would be better if you clarified what you were looking for and why you were interested in it. If you are trying to write a paper on OSS, for example, I would recommend focusing on the latest ones which have garnered a lot of attention and user base, like the Mozilla Suite or OpenOffice.org. --24.147.86.187 16:19, 16 October 2007 (UTC)

Category:Free software contains all Wikipedia's pages on free software (free software and open-source software largely refer to the same thing). The latest versions of free software are available for download from...

• The software authors' websites (examples: [2], [3]). These sometimes offer versions which are still under active development.
• Many Linux distributions can update or install free software using software like synaptic, or directly from the website
• Open-source software sharing sites, like SourceForge host many open source projects
• Your friends who are already using the software may provide you with a copy

So there are quite a few sources for the latest free software --h2g2bob (talk) 16:46, 16 October 2007 (UTC)

Check out Ubuntu and navigate around the nauseatingly stupid disambig page.. to find that Gutsy is coming out in 2 days!! Unfortunately for use gnome-haters, KDE 4 has been pushed back till december. So meh. --froth^t 00:25, 17 October 2007 (UTC)

CSS positioning with Safari

I'm working on a website for a client who happens to be using a mac; I don't have access to a macintosh. While I see a nice gallery page here [4], she sent me a screenshot here: [5]. I've got it in a center tag with relative positioning to shift it from center to provide space; is this not allowed in Safari? example:

<p align="center"><span style="{position: relative; left:-2in}"> <a href="floral.htm">
 <img src="images/thumbs/flowerthumbnail.jpg" width="144" height="96" border="0"></a></span>. 

Kuronue | Talk 17:10, 16 October 2007 (UTC)

I don't know about the Mac/Safari setup - but personally, I deeply distrust anything measured in "inches" because so many people set up the pixels-per-inch thing in their browser incorrectly. And even when they don't - do you really want the images two inches away on an iPhone or a PSP or something? I'd try setting a number in pixels - after all, you set the image dimensions in pixels. SteveBaker 17:21, 16 October 2007 (UTC)

I had it in pixels; she has a very, very large screen at a high resolution and complained that they were too close together (50 pixels of space looked fine on my screen but wasn't showing up on hers) so I tried inches; though, now that I think about it, that might have been the same issue, since it's not showing any space in inches either. Kuronue | Talk 17:27, 16 October 2007 (UTC)

First, I have to note that this would be MUCH easier with a table. But, keeping in the "table are evil/css is good" mindset, you can mimic a table. Wrap all of the images on the top row in a div with width set to 100%. Inside it, have three divs set to a width of 33% with text-align center and display inline. On some browsers, you can use spans - but I've had too many of them refuse to set a width on my span. So, you have three dive, each 1/3rd of the screen and an image inside them comes out centered inside the div. Repeat for the second row but with only two internal divs set to 50% width. -- kainaw™ 17:35, 16 October 2007 (UTC)

Apple recently released a beta of Safari for Windows, so you don't need a Mac to test on Safari any more. You can download it from here. — Matt Eason ^{(Talk • Contribs)} 18:09, 16 October 2007 (UTC)

Sounds like the problem is more to do with this insanely high screen resolution and (possibly) an improperly set up dots-per-inch thing though - I bet Safari has nothing to do with it. The problem is that there is no way (in general) for the computer/graphics-system to know how big the display is - so "inches" is never a meaningful measurement unless every user has gone in and hand-tweaked the settings for that (which hardly anyone ever does!). Using pixels is occasionally useful - and using percentage-of-window-size can help - but the trouble is that people use high res screens in two ways... some people have big monitors with lots and lots of small (but still pretty readable) windows - other people have smaller monitors with one really, really, crisp/sharp window. For the first community you need pixels - for the second window-size-percentages. Basically, we're doomed. SteveBaker 21:03, 16 October 2007 (UTC)

I viewed the website in question in Firefox and Safari at the same resolution (1024x768) and I got the results shown in the screenshot, so the two are definitely rendering it differently. I don't know if either one is at fault or if it's a problem with the code. — Matt Eason ^{(Talk • Contribs)} 21:09, 16 October 2007 (UTC)

The site in question is here [6]. Definately a Safari issue.Changing span to div makes it worse. Any other suggestions? Kuronue | Talk 23:15, 16 October 2007 (UTC)

Try the following...

<html>
<body>
<div style='width:100%;text-align:center;'>
	<span style='display:block;float:left;width:33%;text-align:center;'>FLORAL</span>
	<span style='display:block;float:right;width:33%;text-align:center;'>VINEYARDS</span>
	<span style='text-align:center;'>GRAPES</span>
</div>
<div style='width:100%;text-align:center;'>
	<span style='display:block;float:left;width:50%;text-align:center;'>COASTAL</span>
	<span style='display:block;float:right;width:50%;text-align:center;'>COMMERCIAL</span>
</div>
</body>
</html>

This works for me in Konqueror, which has the same engine as Safari. -- kainaw™ 23:33, 16 October 2007 (UTC)

Ooh, that looks shiny, Kainaw, but I managed to figure it out with help from the apple forums: Safari doesn't like curly braces in its inline CSS tags. Removing them fixed it. Kuronue | Talk 13:40, 17 October 2007 (UTC)

Flash issue exporting to QuickTime

I made a nifty animation in Flash CS3 that I want to export to quicktime so I can import it into Final Cut Pro. I did this MANY times and it worked fine—it exported fine and imported fine. Now, a mere minutes later, when I export it just exports as a black screen. The SWF itself runs fine (it is very code-heavy), but now it won't export at all to quicktime. It's driving me a little mad since I didn't change any of the settings — I just changed some of the code around, and it still works fine in the SWF, but no longer displays (the code changes I made were not major, and I have tried even reverting them to its original state—no dice).

I've tried restarting Flash and everything else numerous times but it seems totally borked. The status log generated by the export shows nothing suspicious. Any thoughts? I'm totally baffled here. --24.147.86.187 19:34, 16 October 2007 (UTC)

OK—now it works again. I changed the export filename, now it suddenly works. WTF. --24.147.86.187 19:37, 16 October 2007 (UTC)

64-bit processors?

Is it true that I need a 64-bit processor to take advantage of computer memory beyond 4 GiB? It makes sense because 2^64 B = 4 GiB, but somehow 32-bit processors still manage to address several tebibytes of hard disk space. If I upgrade to a 64-bit processor, will my 32-bit Linux OS (Fedora 7) and all its programs still work, or will I need a 64-bit version? JIP | Talk 20:00, 16 October 2007 (UTC)

Disks can be broken up into sectors which hold an arbitraty number of bytes. You need only address the sectors, not each individual byte. Still, problems with sector numbering caused a similar problem, not too long ago. Maybe logical block addressing will explain it?

As for 32-bit apps, in general, 64-bit processors are fully backwards-compatible with 32-bit processors. However, you need to use 64-bit code to take advantage of 64-bit features. That is, Fedora 7 will still work, but they still won't be able to reach over 2GB (4Gb?) of memory. --Mdwyer 20:37, 16 October 2007 (UTC)

It's complicated. 2^32 is 4G - so a 32 bit processor can only access 4Gbytes of RAM memory quickly and efficiently. Disk drives don't need to be accessed really quickly so you can access any amount of disk space with a 32 bit processor. There are a few systems out there that use memory bank switching techniques to get more than 4Gbytes of RAM into a 32 bit box - but it is still the case that an individual program can't reach more than 4Gbytes. In some Linux setups, you can get a lot more speed out of your 32 bit processor by using only 2Gbytes of RAM - the reason for which is deep, dark and arcane - and you don't want to know about it!

A 64 bit processor can access over 16,000,000,000,000,000,000 bytes of memory. Which is more than any computer the size of yours will EVER have because there are only about that many atoms in something the size of a RAM chip! So artificial limits like 4Gbyte are gone forever. However, I doubt many motherboards will be able to accomodate the 8 billion or so memory sticks it would take to fully make use of the 64 bits - so there are bound to be limits still.

As previously explained, you have to be running the CPU in full 64 bit mode to take advantage of all of this wonderousness. I'm doing that with SuSE 10.2 on my AMD64-based laptop and it's working very well. I have no idea where Fedora is on this right now. SteveBaker 20:50, 16 October 2007 (UTC)

I think there's always a lot of confusion when referring to a processor or something as "64-bit" or "32-bit". It usually refers to the size of integers the processor uses. Some processors use 64-bits internally, but have a 32-bit interface. Or vice-versa. But the amount of memory a processor can handle is determined by its address bus, not the size of integers it uses. On [7], quoted from [8], it says the address buses usually aren't 64-bit, and it doesn't look like it will ever need to be.

As far as the software question, running 32-bit apps on a 64-bit operating system is usually a tricky (you need 32-bit libraries), but can be done. However, it is usually ok to install the 32-bit operating system on with the 64-bit processor, but there are some issues with that as well (see the webpage linked before.) --Bennybp 21:29, 16 October 2007 (UTC)

Let's not be hasty with the "that's all we'll ever get" notion: $2^{64}$ atoms of silicon is just 0.8 milligrams, so we could have another 16 or so bits to add in even a very small form factor (nearly 4 more if we can use each electron on each atom!), although admittedly 8 of those are taken up by it being bits and not bytes. --Tardis 21:59, 16 October 2007 (UTC)

You know what they say about 640k... --froth^t 00:11, 17 October 2007 (UTC)

Also, I'm very interested to hear why linux has problems with 4GB of memory. I thought this was a limitation of motherboard clocking/timing? --froth^t 00:14, 17 October 2007 (UTC)

It's not a "problem", it's just a matter of making effective use of the 4G virtual address space. If you have 1G of RAM, you can offer 3G of address space to each user process, and keep the 1G of RAM mapped simultaneously so a process can make a system call and the kernel can access whatever it needs without changing the mappings. This is called the "3G/1G split" - 3G for user mode, 1G for the kernel. If you have 2G or 3G of RAM there's the possibility of using a 2G/2G or 1G/3G split, although those options are not widely used. The other main option, called "highmem", allows the kernel to use more RAM than the size of its side of the "split", but since it can't have all that RAM mapped simultaneously, it must do more page table manipulation, causing TLB misses, which slows things down a bit. With highmem, you can use a full 4G of RAM (or even 64G with the PAE extension which has been around since the Pentium Pro). The fact that you can find some tasks that are performed more efficiently with less memory is not because of a "problem" or bug in the big-memory usage, it's just that a certain optimization (mapping all RAM simultaneously and keeping it mapped forever) is not possible when RAM size equals or exceeds the size of the virtual address space. So there's your "deep" answer. (It may also be wrong in a few places. I'm not a kernel programmer and they're the only ones who really understand this stuff.) --tcsetattr (talk / contribs) 06:35, 17 October 2007 (UTC)

Yeah - I think you have that right - but I'm not a (frequent) kernel programmer either! Basically there is a trade-off between the total amount of memory and the amount that one program can use at any one time and between speed of operation through kernel calls. It's really horrible - and that's why we all need to rush out and buy 64 bit processors immediately! I'd really like to use the brain cells that store this useless and annoying crap for something else! SteveBaker 00:20, 18 October 2007 (UTC)

CRW on Linux?

My mother suggested that I finally move from a middle-end digital camera (Canon PowerShot S3 IS) to a DSLR - she has a Canon EOS 10D, she suggested a Canon EOS 400D for me. But the principal issue is, unlike my 100% Windows-only parents, I only use Linux at home. My mother has no problem with converting the CRW files her camera takes in Photoshop to PSD or JPG, but I fear I will not be so lucky. CRW is a Canon proprietary format, and what I have learned, Canon hasn't even learned Linux exists. JIP | Talk 20:04, 16 October 2007 (UTC)

No problem! You can do it in either ImageMajik or GIMP with the right plugins. The low level decoder for these images is a package called dcraw [9] - that page lists about 30 applications that use the library and can therefore read CRW. There is a GIMP plugin that uses UFraw [10] - which in turn uses dcraw - so with some dinking around, you should be good to load CRW into GIMP. This page is a reasonable guide to what you need: [11]. You might want to get your mom to email you some CRW images to try out first before you go out and spend a fortune on the camera. There is a detailed description of the format here if you really want to know what goes on under the hood. There is an article on working with raw camera images up on Linux.com too [12]. SteveBaker 20:41, 16 October 2007 (UTC)

dcraw and the accompanying GIMP plugin were easy enough to install. But I found that when I open a CRW file in the GIMP, all I get is a dialog box with only a few checkboxes for conversion options, and when I click OK, the plugin automatically converts the CRW image into GIMP RGB (which I can then later convert to JPG or PNG or whatever by hand). From what I've seen from my mother's work, Photoshop allows much greater control over the CRW conversion, with a live preview. I'll never get the hang of raw conversion if I can't see by my own eyes which parameter affects what. Also, now I get an error message on stderr from dcraw every time I open a non-CRW file in the GIMP (it still works though). JIP | Talk 17:22, 17 October 2007 (UTC)

Hmmm - well, there's not much I can do about the usability issues - try some other software for doing the conversion? There seemed to be at least 30 of them listed on the dcraw web page[13]. I suppose you could ask whoever wrote the GIMP plugin whether they would consider providing a preview panel - be nice, these guys work for no money! As for the error message - what does it say? Emailing that to the plugin author would probably be useful to him/her too. SteveBaker 00:16, 18 October 2007 (UTC)

Holding on to what I'm doing

Just inquisitive, is there a program out there that can, prettymuch let me fully shut down my computer and then let me go back to the webpages I was on and all the tabs, with the taskbar the way it was before shutdown? I live an overcrowded life, and I'm turning the computer on, turning it off, then having to open up all of what I was doing again. Tried google, but I got a bunch of junk about Applied JavaScript and XHTML. Know of any? I know that it wouldn't work at startup, but would could it be used by clicking a button? YДмΔќʃʀï^→Г_C← 10-16-2007 • 22:16:07

Does hibernating the computer help? I know hibernating is not the same as turning off but for most practical purposes, it could be the same. Firefox lets you quit the browser and open the same web pages next time you open Firefox. (Opera, too, has the feature. Safari and IE7 should also have similar a feature for you to enable.) --Kushal^t 22:22, 16 October 2007 (UTC)

I understand where you're comming from, and I probably will do that, but I can't get back what I was doing, for say last Tuesday when I came out of hibernation could I? I guess I could hibernate though. YДмΔќʃʀï^→Г_C← 10-16-2007 • 22:39:32

Both firefox and windows explorer will open all open windows again if you leave them open when you turn the power button off.--Dacium 23:35, 16 October 2007 (UTC)

Hibernate will restore your computer to exactly the same state it was in when you turned it off (Firefox and any other apps still open and with the same documents/web pages loaded), whether you turned it off yesterday or last Tuesday. As Kushal said, you can also set Firefox to automatically remember what tabs you have open and restore them when you close and re-open it. You can enable this feature by going to Tools > Options... and selecting "Show my windows and tabs from last time" in the very top dropdown menu. — Matt Eason ^{(Talk • Contribs)} 23:44, 16 October 2007 (UTC)

Ok, thanks then! YДмΔќʃʀï^→Г_C← 10-17-2007 • 00:33:45

BTW, OP, your signature is cool. --Kushal^t 19:56, 17 October 2007 (UTC)

Just unicode characters *rolls eyes* Everyone's little brother uses them to make crazy arabic style names in Counter-Strike --froth^t 20:43, 17 October 2007 (UTC)