Wikipedia:Reference desk/Archives/Computing/2009 January 11

= January 11 =

nature of ARP traffic
For the first time today my network traffic monitor is showing ARP requests. In one minute it shows some 1500–2000 ARP frames. Is this normal background network activity, and why would I be seing it in Wireshark or Microsoft Network Monitor only now? (I did get a new cable modem a few weeks ago.) –Outriggr § 01:45, 11 January 2009 (UTC)


 * 1500-2000 ARP frames per minute is only normal if your host is looking up (or being looked up by) 750-1000 other hosts on the same subnet (i.e. subnet==hosts whose (ip & mask) match). -- Fullstop (talk) 01:56, 11 January 2009 (UTC)


 * My IP address is not involved, if that's what you mean: what I see is "CCC.CCC.vvv.vvv asks for CCC.CCC.vvv.vvv", where C is fixed and v is variable. –Outriggr § 02:03, 11 January 2009 (UTC)


 * You should not be seeing that traffic unless your 'CCC.CCC' matches the 'CCC.CCC' you are seeing. And... if your cable modem is a router (not a PPPoE thingie), you should be on on your private 192.168.x.x (or similar) net, and not be seeing any traffic at all. -- Fullstop (talk) 02:14, 11 January 2009 (UTC)


 * Well, my C matches their C. I don't think my modem is a router. What I don't get—now that you've implied maybe this is normal—is why I have seen this traffic for the first time only now, and in the bigger scheme of things, why this traffic exists—is it the result of more unnecessary Windows services? Why would computers on my subnet (if that's the right term) need to know the hardware address of others? It's little to do with the common uses of residential internet service, no? Thanks BTW. –Outriggr § 02:24, 11 January 2009 (UTC)


 * A little background info first:
 * The MAC (ethernet) level, which includes your network card, knows nothing about IP. Accordingly, the ethernet packets (as opposed to the IP packets, which are encapsulated inside the ethernet ones), need to be tagged with the ethernet address of the recipient. So, to look up the ethernet address of another host, your machine does a hardware-level broadcast that asks "Is IP address x.x.x.x yours?" Other hosts on your net see these requests, and if the x.x.x.x matches theirs, they respond with their ethernet address so that both sides then know how to translate IP addresses to ethernet addresses for further communication.
 * Now, it would be extremely inefficient if, for every single IP packet sent, each side had to first ask for the hardware address of the other party. So, ethernet addresses are cached, and a new "Is IP address x.x.x.x yours?" request will only be sent when the cache entry turns stale (typically about 10 minutes after communication with that destination ceases).
 * When ARP caching is working, and there are no malicious hosts around, it is not at all normal to see 1500-2000 ARP frames per minute. Even in a big LAN party, the ARPS that will necessarily precede regular communications are never seen again.
 * I can only imagine a handful of reasons why anyone would see the kind of ARP traffic that you are seeing, and most of those reasons are not good. Presumably, you are seeing ARP requests (the "Is IP address x.x.x.x yours?" broadcasts) coming from only a single host (or, by now, a handful of them). Such a single-origin pattern could indicate one of two things: A) The gateway (on the border across which all internet traffic is flowing) was misconfigured such that wasn't caching properly. B) There is a bad apple in your midst that is probing for security loopholes on the other machines on the subnet, or trying to poison their ARP caches.
 * Tip: Browse/edit over https if you can. In a non-switched environment where you can see other people's traffic, they can also theoretically also see yours, and so can nick your password. -- Fullstop (talk) 12:28, 11 January 2009 (UTC)

Good explanation. Perhaps I should have looked at the traffic more closely. Over less than four minutes there were the following number of ARP frames from these IPs: aaa.bbb.0.1	508 aaa.bbb.100.1	735 aaa.bbb.12.1	670 aaa.bbb.16.1	632 aaa.bbb.20.1	539 aaa.bbb.24.1	916 aaa.bbb.28.1	523 aaa.bbb.4.1	688 aaa.bbb.8.1	604 aaa.bbb.96.1	455

So there's a clear pattern... are dot-1 machines gateways? (I once began an introductory networking course - suffice it to say, hanging around with just-out-of-high-school students in basic courses and 3-hour no-content computer labs while losing pay was not how I wanted to learn IT. ;) –Outriggr § 23:18, 11 January 2009 (UTC)


 * a) Yes, .1s are typically (but not necessarily) gateways. Mine for instance is at .254.
 * b) I have a very strong feeling that those gateways are misconfigured, most likely with inappropriate netmasks. Those hosts are very obviously chatting on the subnets of their neighbours, and not compartmentalizing (what you show as) a /17 network (32,768 addresses) is just ... bad.
 * Such a large subnet is a very possible cause for lots of ARP traffic. Not just because there are so many hosts to talk to, but also because it can cause the caches to be continuously purged for lack of space ("ARP cache thrashing" i.e. the cache entries are discarded to make space for new ones, even though the old ones aren't yet stale). -- Fullstop (talk) 02:43, 12 January 2009 (UTC)


 * So what you're saying is that you want to work for my ISP? :) –Outriggr § 04:33, 12 January 2009 (UTC)

svg 2 png curious
When I right click to save a (*.svg) graphic, it seems I can only save as *.png. I'm curious as to why.


 * OS .. Vista 64
 * Firefox, Chrome, and IE

Is it because I don't have an svg editing program installed? Ched (talk) 06:10, 11 January 2009 (UTC)
 * wait ... I do have GIMP 2.6 installed, so I guess that's not it Ched (talk) 06:13, 11 January 2009 (UTC)
 * [[Image:Hammer_and_sickle_on_a_red_star.svg|200px|right|thumb|the example]]No, that's not your problem. Let me explain. MediaWiki has SVG capabilities, but most/all browsers don't. So MediaWiki has to convert the SVG into a PNG file for display in a browser. In order to download an SVG file, you can follow the link to the image description page, then click on the big picture and click "Save link as..." (Note: That's not "Save image as...). i.e., if you wanted to download Image:Abusive admin star.svg (to the right), you could right-click that image and click "Save image as..." but that would give you a PNG. So don't do that. Instead, left-click the image to go the the image description page (Image:Abusive admin star.svg). There, right-click the big image and click "Save link as..." which will give you the SVG. (wow that was a long answer to a short question...) flaminglawyerc 06:41, 11 January 2009 (UTC)


 * Cool, thx and it works. Now I really feel bad I can't be more help with the HD... still thinking though ;)  Ched (talk) 06:51, 11 January 2009 (UTC)


 * Now that's weird... The image to the right (as provided by Flaming) appears as an empty box in my browser, yet it appears as its proper image if I click on it or if I see it on other pages. I'm using IE7 and Vista if that is of interest.  Astronaut (talk) 11:44, 11 January 2009 (UTC)


 * Same here (Firefox 3.0.5 and Xp). --NorwegianBluetalk 13:49, 11 January 2009 (UTC)
 * Yeh... same here, I thought it was just me. I'm on Firefox 1, Damn Small Linux (don't ask, please). flaminglawyerc 21:22, 11 January 2009 (UTC)
 * That'll be because the image above is not an svg, but a sort of pseudo-png (/blahdeblah.svg.png). Because my browser (among many others) has SVG support, when you actually get through to the file it works. It's the conversion from svg to png that the wiki software's doing that messes it up. - Jarry1250 (t, c) 21:31, 11 January 2009 (UTC)
 * What's with the 'width="300pxsdfsdfsdf"' in the .svg source? If I remove 'sdfsdfsdf' the image moves to the left of the page. Maybe the white box is just the left side of the image. &mdash;Noah 22:54, 11 January 2009 (UTC)
 * This looks fixed to me, how about for you? Hammer_and_sickle_on_a_red_star.svg &mdash;Noah 23:02, 11 January 2009 (UTC)
 * And now it's fixed. Well spotted Noah.  Astronaut (talk) 15:50, 12 January 2009 (UTC)

let's try this again...
In my previous post, i gave way too much information and expected like 5 answers. So let's start with only one question, and I'll cross any more bridges when I come to them.

I've been messing around with my IDE HDD (like taking it out, putting it back in, etc.). Now my comp won't boot off of it. My BIOS's boot order is in the table to the right. When I turn on my comp, it skips all the way down to "network boot," which is weird because I have an OS installed on my HDD, which means it should boot from it. I know my BIOS doesn't have anything wrong with it because I can still boot from my Linux LiveCD's.

I didn't have my harddrive near anything magnetic. All I did was take it out, put it onto a desk, and put it back in. What's wrong with my harddrive, and iss there anything I can do to get this HDD to boot again? flaminglawyerc 06:30, 11 January 2009 (UTC)


 * I'm assuming you have both power and ide ribbon plugged in, checked to make sure you didn't pull ide cable lose from mobo. could you have knocked off the jumper that makes drive boot Master, CS, or slave?  And when you say not near magnetic, I assume you include electrical.  ...(thinking) .... does the boot LiveCD recognise the drive as there and with info on it? Ched (talk) 06:37, 11 January 2009 (UTC)
 * Wow. I forgot to check the mobo. Checking... (this might take a while, my case is uber-tiny and has like 700 cords in like 2 cubic inches) [I need a new comp so bad :( ] flaminglawyerc 06:49, 11 January 2009 (UTC)
 * And that's what I find to be the biggest downside to a Small form factor. Useight (talk) 08:09, 11 January 2009 (UTC)
 * Yes, the biggest, but certainly not the only... :( flaminglawyerc 08:25, 11 January 2009 (UTC)


 * When you are tugging cables, especially in a small space, it is easy to pull something else out by accident. I suggest you go round making sure all the plugs are properly seated in their sockets.  Try to get your PC back to how it was before you started messing with the drives at all, including the master/slave jumpers, power connectors and IDE cables.  You should then be confident in finding out whether or not you have broken your old hard drive.  Astronaut (talk) 12:34, 11 January 2009 (UTC)


 * Another possibility is that a static electricity discharge to any exposed circuits/wires on the hard drive could have damaged the drive electronics. Have you experienced static buildup recently, and do you recall shocking/being shocked by the drive? —Preceding unsigned comment added by 74.137.108.115 (talk) 15:27, 11 January 2009 (UTC)


 * If static buildup is causing you problems, try new and improved "Bounce" ... I'm sorry, I just couldn't resist that ;) ... just hand me my wiki-demerit on my way out ;) ... Ched (talk) 17:19, 11 January 2009 (UTC)


 * One possibility is that the data cable is plugged in backwards, as some of them actually aren't keyed to prevent this. Another possibility is that you plugged the hard drive into another cable or a different position on the same cable.  If you do this you often need to change the jumper position to indicate the new "location" of the hard drive (primary master, primary slave, secondary master, or secondary slave). StuRat (talk) 06:59, 12 January 2009 (UTC)


 * I've never seen jumpers on a hard drive for primary/secondary (with the exception that some drives use primary/secondary as PC replacements for master/slave). From the drive's perspective, it's connected to an IDE bus with no visibility to other parallel IDE busses. Also, some motherboards have more than two IDE busses, which would presumably confuse a drive that was somehow capable of identifying which IDE bus it was on. --74.137.108.115 (talk) 08:50, 12 January 2009 (UTC)

Anyone familiar with the development of free software?
Now I know many people develop free software and then often does some revolutionary process or what not that if the software was made by a company that sold it, that company would patent such a process. Now as free software is often written by people who are small groups or even one persn and don't have the money to engage in legal battles, well I'd think that rich companies would come along all the time and just steal their idea patent them and claim it as their own. And some software processes can be patented, such as eBay patented its "Best Match" search system. Now I know of one instance where a big company ripped off a project called KittenAuth, see this google search query for more info. Now I know keeping a patent costs as much as a house each decade in all the various fees and most people who develop free software don't do any intellectual property stuff to it, so I'm surprised that big companies aren't constantly stealing from freeware/open source work an then patenting it and then forcing the true inventors from working on their ideas because the true inventors can't afford the lawyers. Well anyway, anyone know more about this related subject? I think there's a bunch of people on Wikipedia who develop free open source software and so would know. Are you ready for IPv6? (talk) 08:30, 11 January 2009 (UTC)


 * One does not patent software in the sense of programs. Patents are issued for means to get a certain job done in a particular fashion, i.e. algorithms. Software is artwork, and like any other creative act, is (automatically!) protected by copyright. Unlike a patent, a copyright costs nothing (unless you need to defend it in court, in which case help from the FSF and pro-bono council is available).
 * As you point out, patents are expensive. Consequently, engineers (software or otherwise) do not patent algorithms unless there is money expected in return.
 * And there is always more than one way to skin a cat... png would probably not exist if the Lempel-Ziv-Welch algorithm had not been patented.
 * -- Fullstop (talk) 10:44, 11 January 2009 (UTC)


 * In the United States we currently have a first-to-invent system and not a first-to-file system. That makes it reasonably hard to "steal" patents from projects that involve multiple witnesses, open-source, and lots and lots of timestamps. If it changes to first-to-file, as it is likely to do so at some point, that would make it a lot easier.
 * But could it happen? Sure, it could. Big Company X could rely on the idea that the open source people aren't organized enough to file a lawsuit. Would that be true? Almost surely not—there are plenty of organizations (EFF, FSF, etc.) that would love to sue some big, thieving company on behalf of open source software, as the publicity would be great and there could be real money involved. So it's not as risk-free a venture as you might think. --98.217.8.46 (talk) 15:11, 11 January 2009 (UTC)


 * Yep - it's called "prior art" - if you can prove that you used something before someone else filed a patent on it - then the patent is invalid and may simply be ignored. That's why nobody has patented the wheel (for example).  Software engineers who produce OpenSource code are generally also very much against software patenting - so they are very unlikely to want to patent something themselves and typically have no problem at all with a commercial organisation "using their ideas"...in fact, part of the idea of OpenSourced software is to promulgate ideas. SteveBaker (talk) 01:19, 12 January 2009 (UTC)


 * If your opensource software was stolen, and you didn't have the money to mount a legal effort, there's a chance you might convince the Free Software Foundation to help you out. APL (talk) 16:13, 12 January 2009 (UTC)

Weird file format
Every time I try to save a .exe file now it turns into an EFW file and it wont open what do I do? --124.254.77.148 (talk) 11:41, 11 January 2009 (UTC)
 * With what software are trying to save the .exe file? Astronaut (talk) 11:45, 11 January 2009 (UTC)
 * This link might provide some explanation. Astronaut (talk) 11:47, 11 January 2009 (UTC)

GCC argument order
I can do

$ gcc -o bisoncalc bisoncalc-calc.o bisoncalc-parser.o bisoncalc-lexer.o -Wall -ly -lfl -g -O2

and get a working bisoncalc executable. Or then I can swap the arguments:

$ gcc -Wall -ly -lfl -g -O2 -o bisoncalc bisoncalc-calc.o bisoncalc-parser.o bisoncalc-lexer.o bisoncalc-calc.o: In function `main': /foobarpath/bison_calc_autotools/src/calc.c:7: multiple definition of `main' /usr/lib/gcc/x86_64-linux-gnu/4.3.2/../../../../lib/liby.a(main.o):(.text+0x0): first defined here bisoncalc-parser.o: In function `yyparse': /foobarpath/bison_calc_autotools/src/parser.c:1391: undefined reference to `yyerror' /foobarpath/bison_calc_autotools/src/parser.c:1537: undefined reference to `yyerror' bisoncalc-lexer.o: In function `yylex': /foobarpath/bison_calc_autotools/src/lexer.c:850: undefined reference to `yywrap' collect2: ld returned 1 exit status

The *.o files were generated from Makefile.am:

AM_YFLAGS = -d BUILT_SOURCES = parser.h bin_PROGRAMS = bisoncalc bisoncalc_SOURCES = calc.c parser.y lexer.l bisoncalc_CFLAGS = -Wall -ly @LEXLIB@

Unfortunaly when I run make it seems to prefer the form that fails. Why it fails and what can I do to fix it? This is about the first time I have used automake and I'm a bit lost. --194.197.235.38 (talk) 15:15, 11 January 2009 (UTC)

Ok I should have used google for 30 minutes longer before asking and I would have found out linker flags should be done with _LDADD. --194.197.235.38 (talk) 15:45, 11 January 2009 (UTC)

There is any free converter that is able to convert from .wav to .aiff
there is any free converter thats convert from .wav to .aiff?? —Preceding unsigned comment added by 201.79.35.36 (talk) 16:49, 11 January 2009 (UTC)
 * I don't know if Audacity can do it, but it's an excellent program and this would be right up its alley. Useight (talk) 17:15, 11 January 2009 (UTC)


 * Yes, it will export to 8 and 16 bit PCM AIFF and 32 bit float AIFF. 87.114.7.226 (talk) 18:13, 11 January 2009 (UTC)

web sites from a search engine
when a search engine brings up sites for you to explore, where does the information above the web site come froom and how does it get to appear on the web site —Preceding unsigned comment added by 70.213.3.97 (talk) 21:46, 11 January 2009 (UTC)


 * (I am not quite sure that I have understood your question.) When you open a web page from Google or any similar search engine, sometimes the upper part of the web brower page is filled with content from the web engine and not from the web page you just opened. This may be the case if you for example are searching for images, for then Google may want to remind you about the copyright laws. Technically, Google is not sending your browser to the desired web page (e.g. www.example.com), but to a frameset web page on google.com. This frameset web page consists of two frames, the uppermost displaying informaiton from the search engine, and the large one displaying the actual web page (e.g. www.example.com). --Andreas Rejbrand (talk) 21:54, 11 January 2009 (UTC)

No. what I am trying to get to is the list of sites the search engine gives you to select from what is the information called or the subject matter called and where does it come from and how does it get there to be able to get to a web site? for example: BEST PRODUCT, can't find anything better on the web, great stuff, nothing like it. www.example.com/html Those words beginning with "BEST" and ending with "it".


 * OK. In a HTML page's HEAD section, you can add meta tags. Consider the following sample HTML document:


 * The META tags, particularly  is used by the search enging to describe the web page in the engine's hit list. --Andreas Rejbrand (talk) 13:18, 12 January 2009 (UTC)


 * Google calls them snippets. Typically they're either a chunk of the page showing your query in context, with the query in bold (eg on a search for, the second result is "30 Dec 2008 ... Reference desk — Serving as virtual librarians, Wikipedia volunteers tackle your questions on a wide range of subjects. ..."), the meta description, or a description from DMOZ. Matt Cutts talks about the anatomy of a search result here — Matt Eason (Talk &#149; Contribs) 13:32, 12 January 2009 (UTC)

creative zen and bloody vista
hi all, don't get me wrong i've never really had any problems with vista, but... santa brought me a second hand creative zen nomad jukebox xtra for xmas, santa should have researched a bit more before he ordered it from e-bay, but santa was "tired" after a hard day mixing with Mr daniels, sooooo.... vista no likey me zen, it no see it connected, it driving me insane. have been on various sites, saw and tried various fixes, have came to realise that i'm not the only one that has this problem... and here is the rub dear friends, can the ref desk of St Wiki fix the unfixable? can it come to the aid of a broken man? please say it's so, please let me listen to the greatest hits of whitesnake in the comfort of my zen moment... Here i go, again on my own/Down the only road that I have ever known...Perry-mankster (talk) 21:51, 11 January 2009 (UTC)


 * You haven't actually listed the specifics of the problem. It's possible that the new computer doesn't meet the minimum req's for Vista, in which case a less bloated O/S, like Windows XP or Linux, may still work. StuRat (talk) 06:49, 12 January 2009 (UTC)
 * Have a look at this link it seems that additional downloads are required to work. BigDunc  Talk 15:24, 12 January 2009 (UTC)


 * aah bigdunc have been to this site already, scooted over to the free download last night, waited the 20 minutes it took to download and then watched as vista repeatly told me that the program would not install. The main prob sturat is that the zen is not compatable with vista (or vice versa), vista does not recognise that the device is there and closes down the music management software if you try and load tracks onto the zen (which it says isn't there) windows xp can work (with a bit of jiggery pokery), but hell i've got vista I WANT IT TO WORK GODDAMN IT! sorry had to get that oot ma system, anyhoo there is apparently a few 'fixes' out there, but they are a bit hit and miss wether they work or not, i was hopeing the ref desk might know of a surefire, 100% 'that'll work bigman' way to zen enlightment Perry-mankster (talk) 18:02, 12 January 2009 (UTC)
 * I recal somewhere that Windows Media Player 11 is causing problems, have you tried uninstalling it and adding back an older version? Or just take it off completly. BigDunc  Talk 18:32, 12 January 2009 (UTC)
 * thanx for the suggestions bigdunc, have found a couple of 'fixes' that might work, so will try them, thanx again Perry-mankster (talk) 21:14, 15 January 2009 (UTC)

foobar2000 songs not in order
In my foobar2000 for some artists the tracks are not in order and are mixed up (ex. track 1 followed by track 4). I am using panelsUI as well. Can someone help me identify the problem? —Preceding unsigned comment added by 154.20.167.198 (talk) 23:44, 11 January 2009 (UTC)


 * What criteria is foobar2000 using to sort songs? This can most likely be changed in a menu or by clicking on a relevant table heading somewhere. You should also compare the files from a folder that sorts correctly to one that doesn't; my guess would be that foobar2000 is sorting by filenames, and the "correct" filenames are of the pattern "(ARTIST) TRACK# TITLE" while the "incorrect" filenames omit the TRACK#, thus sorting alphabetically by title. If your files have additional information (mp3s for instance have a Track# field in their header, though it isn't reliably used) foobar2000 may allow you to sort the albumn using that field. Otherwise, you could rename the files such that their track number is included in their filename. —Preceding unsigned comment added by 74.137.108.115 (talk) 00:33, 12 January 2009 (UTC)

.bat file crash
I'm making a .bat file, but I have a problem: when a user inputs a space (" "), it just crashes, instead of saying "Option not valid", here is the code:

@echo off echo 1 - Print "Option 1" echo 2 - Print "Option 2" echo Option: set /p option= if ‘%option%’ == ‘1’ goto OPT_1 if ‘%option%’ == ‘2’ goto OPT_2 goto error :OPT_1 echo Option 1 goto end :OPT_2 echo Option 2 goto end :error echo Option not valid :end pause

Can anyone help? Thanks. SF007 (talk) 23:57, 11 January 2009 (UTC)
 * I don't have a Windows machine to hand to test it out, but I think that the variable being blank is a problem. The solution might be to code as follows:

if ‘x%option%’ == ‘x1’ goto OPT_1 if ‘x%option%’ == ‘x2’ goto OPT_2
 * You'll see that I've inserted the character 'x' twice in each line so the result is never blank.- gadfium 01:16, 12 January 2009 (UTC)


 * Try this:
 * @echo off
 * if %1.==1. goto opt_1
 * if %1.==2. goto opt_2
 * if not %1.==. echo "%1" is not a valid option.
 * echo 1 - Print "Option 1"
 * echo 2 - Print "Option 1"</tt>
 * <tt>set /p option=Option:</tt>
 * <tt>call %0 %option%</tt>
 * <tt>goto end</tt>
 * <tt>:opt_1</tt>
 * <tt>echo Option 1</tt>
 * <tt>goto end</tt>
 * <tt>:opt_2</tt>
 * <tt>echo Option 2</tt>
 * <tt>goto end</tt>
 * <tt>:end</tt>
 * That's off the top of my head, but should work. -- Fullstop (talk) 01:24, 12 January 2009 (UTC)


 * Hey gadfium, good idea, but it did not worked. Your solution worked Fullstop (I only had to add pause at the end), although it is a bit complicated for me to understand... I think the program calls itself right? Can you explain the theory behind it? SF007 (talk) 03:26, 12 January 2009 (UTC)


 * a) Yes, it calls itself, thus making use of the fact that arguments passed on the command line are stripped of extraneous whitespace. As a side-effect, it also makes the script usable as "your.bat 1", etc, which is the way things are usually done in the command line world.
 * b) With respect to "%1", etc: "%0" is the name of the script itself. "%1" is the first argument, "%2" the second, and so on. Variables (==Environment variables) need to be read as %variable% when they are created with the <tt>set</tt> command. In contrast, "%1" etc are not variable; the cannot be assigned new values (except with <tt>shift</tt>). And see also <tt>for</tt>, the "variables" of which need to be addressed as %%scalar%% (double % on both sides) and also cannot be assigned new values.
 * c) "Option:" is written with ':' because that's how you had previously written it. -- Fullstop (talk) 03:44, 12 January 2009 (UTC)
 * Thanks a lot mate SF007 (talk) 23:44, 12 January 2009 (UTC)